Which of these is NOT a method of communication?
(b) Encryption The three methods are verbal, non-verbal and visual.
Minimum Learning Support Material · CBSE 2026-27
Notes · Numericals with working · 1026 solved questions · Flashcards · Quiz · Practice Paper Maker
Marks, pattern and blueprint for Class XI AI (843), session 2026-27
| Unit | Name | Theory hrs | Practical hrs | Marks |
|---|---|---|---|---|
| 1 | Introduction: AI for Everyone | 4 | 10 | 4 |
| 2 | Unlocking your Future in AI | 6 | 10 | 5 |
| 3 | Python Programming | 10 | 20 | 5 |
| 4 | Introduction to Capstone Project | 6 | 15 | 5 |
| 5 | Data Literacy - Data Collection to Data Analysis | 6 | 15 | 6 |
| 6 | Machine Learning Algorithms | 9 | 15 | 6 |
| 7 | Leveraging Linguistics and Computer Science | 5 | 10 | 5 |
| 8 | AI Ethics and Values | 4 | 5 | 4 |
| Part B total | 50 | 100 | 40 | |
| Part A: Employability Skills (2 marks per unit) | 10 |
| Component | Marks |
|---|---|
| IBM SkillsBuild / other industry certification | 5 |
| Capstone project | 12 |
| Bootcamp / internship / other start-ups | 7 |
| Practical file | 10 |
| Lab test / written exam (based on the practical file) | 10 |
| Viva voce (practical file and project) | 6 |
CBSE publishes a sample paper for Class XII (843) only. Schools set the Class XI paper on the same design, so this kit follows that pattern with the Class XI units.
| Question | What comes | Attempt | Marks |
|---|---|---|---|
| Q1 | 6 objective questions on Employability Skills | any 4 | 1 x 4 = 4 |
| Q2 to Q5 | 6 objective questions each from the eight AI units (MCQ, True/False, Assertion-Reason, small numericals, Python output) | any 5 in each | 4 x 5 = 20 |
| Q6 to Q10 | Employability Skills, 20 to 30 words each | any 3 of 5 | 2 x 3 = 6 |
| Q11 to Q16 | AI units, 20 to 30 words each (may be a numerical or Python output) | any 4 of 6 | 2 x 4 = 8 |
| Q17 to Q21 | AI units, 50 to 80 words each: Python program, design thinking case, statistics/matrix numerical, ML numerical, NLP case | any 3 of 5 | 4 x 3 = 12 |
Each cell shows the number of questions given from that unit. Every unit gets three one-mark questions; the long questions cover the practical-heavy units.
| Unit | 1mark | 2marks | 4marks | Totalquestions |
|---|---|---|---|---|
| Part ACommunication Skills-III | 1 | 1 | - | 2 |
| Part ASelf-Management Skills-III | 2 | 1 | - | 3 |
| Part AICT Skills-III | 1 | 1 | - | 2 |
| Part AEntrepreneurial Skills-III | 1 | 1 | - | 2 |
| Part AGreen Skills-III | 1 | 1 | - | 2 |
| Part BUnit 1: Introduction: AI for Everyone | 3 | 1 | - | 4 |
| Part BUnit 2: Unlocking your Future in AI | 3 | 1 | - | 4 |
| Part BUnit 3: Python Programming | 3 | 1 | 1 | 5 |
| Part BUnit 4: Introduction to Capstone Project | 3 | 1 | 1 | 5 |
| Part BUnit 5: Data Literacy - Data Collection to Data Analysis | 3 | 1 | 1 | 5 |
| Part BUnit 6: Machine Learning Algorithms | 3 | - | 1 | 4 |
| Part BUnit 7: Leveraging Linguistics and Computer Science | 3 | - | 1 | 4 |
| Part BUnit 8: AI Ethics and Values | 3 | 1 | - | 4 |
| Questions given | 30 | 11 | 5 | 46 |
| To be answered | 24 | 7 | 3 | 34 |
| Marks | 24 | 14 | 12 | 50 |
Where the 50 marks come from, easiest first
10 marks · five units, 2 marks each · detailed chapters
Communication is the exchange of information, ideas and feelings so that the receiver understands exactly what the sender means. At Level III we go deeper into methods, communication styles and writing skills - the skills an employee uses daily to write e-mails, attend meetings and deal with customers.
| Oral (spoken) | Written | |
|---|---|---|
| Examples | meetings, phone calls, presentations, interviews | letters, e-mails, reports, notices, chat |
| Merits | fast, personal, instant feedback, tone adds meaning | permanent record, can be planned and edited, legal proof |
| Limits | no record, may be forgotten, unplanned words | slower, needs good language, feedback delayed |
Non-verbal cues often tell more than words.
happiness, surprise, anger, confusion
waving, pointing, thumbs-up
upright = confident; slouched = bored
confidence, interest, honesty
handshake, pat on the shoulder
personal distance in conversation
tone, pitch, volume, pace, pauses
dress and grooming send messages too
| Personal space zone | Distance (approx.) | Used for |
|---|---|---|
| Intimate | up to 0.5 m | family, close friends |
| Personal | 0.5 to 1.2 m | friends, conversation |
| Social | 1.2 to 3.5 m | colleagues, customers, interviews |
| Public | more than 3.5 m | speeches, lectures |
Infographics, charts, graphs, maps, signs, icons, logos and videos. They make data easy to understand: a sales graph shows growth faster than a table of numbers.
clear, honest, respectful; "I" statements; listens
loud, dominating, blaming; ignores others' feelings
hides anger; sarcasm, sulking, silent treatment
gives in, avoids conflict, ignores own needs
| Situation: a colleague takes credit for your work | Response |
|---|---|
| Submissive | says nothing, feels hurt |
| Aggressive | "You are a cheat!" in front of everyone |
| Passive-aggressive | "Congratulations on *your* great idea," with sarcasm |
| Assertive | "I prepared the data for this report. Please mention my part when you present it." |
"I feel ... when ... because ... I would like ..." - e.g. "I feel worried when the report is late because the client is waiting. I would like it by 4 p.m."
A sentence is a group of words with a subject and a verb that gives complete sense: "The bell rang." A phrase is a group of words without complete sense and usually without a subject-verb pair: "after the bell", "a very tall building".
| Part | Meaning | Example: *Riya wrote a poem.* |
|---|---|---|
| Subject | who or what the sentence is about | Riya |
| Predicate | what is said about the subject (verb + rest) | wrote a poem |
| Object | receives the action | a poem |
| Kind | Purpose | Example |
|---|---|---|
| Declarative | statement | The meeting is at 3 p.m. |
| Interrogative | question | Have you sent the report? |
| Imperative | command / request | Close the door, please. |
| Exclamatory | strong feeling | How lovely the garden is! |
| Part | Function | Example |
|---|---|---|
| Noun | name | Kavya, Mumbai, team, honesty |
| Pronoun | replaces a noun | she, it, they, who |
| Verb | action or state | build, is, seem |
| Adjective | describes a noun | urgent task, three files |
| Adverb | describes a verb / adjective / adverb | quickly, very, always |
| Preposition | relation of place, time, direction | at, on, in, under, towards |
| Conjunction | joins | and, but, because, although |
| Interjection | sudden feeling | Oh! Alas! Bravo! |
| Article | Use | Examples |
|---|---|---|
| a | before a consonant sound | a book, a university (yoo sound), a one-rupee coin |
| an | before a vowel sound | an apple, an hour (silent h), an MBA |
| the | a particular or unique thing | the Sun, the report you sent, the Ganga |
The choice of a/an depends on the sound, not the letter: *an honest man*, *a European city*.
Topic sentence: Regular exercise improves students' performance. Supporting: It increases blood flow to the brain, which improves concentration. It reduces stress before exams. Students who play sports also learn teamwork. Conclusion: Therefore, schools should include at least 30 minutes of physical activity every day.
Tips: one idea per paragraph, use linking words (firstly, moreover, however, therefore), keep sentences short, check spelling and punctuation.
A sentence has a subject and a verb and gives complete meaning, e.g. "The train arrived late." A phrase is a group of words without complete meaning, e.g. "on the platform".
Assertive - clear and respectful ("Please let me finish my point"). Aggressive - loud and blaming ("You never listen!"). Passive-aggressive - indirect anger or sarcasm ("Fine, do whatever you like"). Submissive - gives in and hides own needs ("It's okay, whatever you say"). Assertive is the most effective style.
Self-management means taking charge of our own behaviour, appearance, learning and time. At this level we study personal grooming, self-exploration, teamwork and time management - the habits that make a person job-ready.
Our appearance creates the first impression - in an interview, the first few seconds matter. Grooming is taking care of personal cleanliness and dressing appropriately for the occasion.
daily bath, deodorant, clean teeth
clean, trimmed and neatly combed
short and clean
clean, ironed, suitable for the place
polished and clean; matching socks
stand straight, smile, make eye contact
| Occasion | Appropriate dress |
|---|---|
| School | clean, ironed uniform, ID card, polished shoes |
| Interview | formal: shirt and trousers / salwar suit or sari, formal shoes |
| Workplace (office) | business formal or business casual as per rules |
| Factory / lab | safety gear: apron, gloves, helmet, safety shoes |
| Party | smart casual |
Grooming is about being clean, neat and appropriate, not about expensive clothes.
Self-exploration is discovering who we are - our interests, abilities, values, personality and goals - so that we can choose the right career and grow.
| Method | How it helps |
|---|---|
| Reflection and journaling | writing daily thoughts shows patterns in our behaviour |
| Feedback from others | teachers, parents and friends see things we miss |
| Interest and personality tests | suggest suitable careers |
| Trying new activities | clubs, sports, volunteering reveal hidden talents |
| SWOT analysis | strengths, weaknesses, opportunities, threats |
Kabir thought he liked only computers. After volunteering at a school event, he found he enjoyed organising and talking to people. A career test showed strengths in both technology and communication. He now plans to become a product manager - a role that mixes both.
A team is a group of people working together towards a common goal. Most jobs require teamwork - surgeons, cricket teams, software teams, event managers.
everyone knows what to achieve
each member knows their job
members rely on each other
open and regular sharing
guides, motivates and decides
appreciating everyone's effort
| Factor | Meaning |
|---|---|
| Cooperation | helping one another |
| Respect | valuing others' views and differences |
| Sharing credit | "we did it", not "I did it" |
| Accountability | owning one's part of the work |
| Conflict resolution | discussing disagreements calmly |
| Urgent | Not urgent | |
|---|---|---|
| Important | Do first: project due today, exam tomorrow | Schedule: revision plan, exercise, skill learning |
| Not important | Delegate / minimise: some calls, routine requests | Eliminate: aimless scrolling, gossip |
Time management tips:
5:00-6:00 games; 6:00-7:30 study the toughest subject; 7:30-8:00 dinner; 8:00-9:00 homework; 9:00-9:30 revise the day's notes; 9:30-10:00 relax, no phone in bed. The plan has priorities, breaks and enough sleep.
Clear goal, defined roles, trust, open communication, good leadership and recognition of effort. (Any four.)
Personal grooming means keeping the body clean and dressing neatly and suitably. Checklist: bath and clean teeth; trimmed hair and nails; clean, ironed formal clothes; polished shoes; minimal accessories; carry documents in a file; upright posture and smile.
A word processor is application software used to create, edit, format, save and print documents such as letters, reports, resumes and notices. Examples: LibreOffice Writer (free and open source, saves .odt), Microsoft Word (.docx) and Google Docs (online).
insert, delete, move text anytime
fonts, colours, alignment, styles
finds spelling and grammar mistakes
reuse text quickly
add images, tables and charts
save as PDF, e-mail, print
| Part | Use |
|---|---|
| Title bar | shows the document name |
| Menu bar | File, Edit, View, Insert, Format, Styles, Table, Tools |
| Standard toolbar | New, Open, Save, Print, Undo |
| Formatting toolbar | font, size, bold, alignment, bullets |
| Ruler | margins and indents |
| Work area | where we type |
| Status bar | page number, word count, language, zoom |
| Task | Shortcut / menu |
|---|---|
| New document | Ctrl + N |
| Open | Ctrl + O |
| Save | Ctrl + S |
| Save As (new name or format) | Ctrl + Shift + S |
| Ctrl + P | |
| Export as PDF | File > Export as PDF |
| Close | Ctrl + W |
| Action | Shortcut |
|---|---|
| Select all | Ctrl + A |
| Copy / Cut / Paste | Ctrl + C / X / V |
| Undo / Redo | Ctrl + Z / Ctrl + Y |
| Find | Ctrl + F |
| Find and Replace | Ctrl + H |
| Delete a character to the left / right | Backspace / Delete |
| Alignment | Shortcut | Use |
|---|---|---|
| Left | Ctrl + L | normal text |
| Centre | Ctrl + E | headings, titles |
| Right | Ctrl + R | date, address in letters |
| Justify | Ctrl + J | neat text with straight edges on both sides (books, reports) |
Other paragraph settings: line spacing (single, 1.5, double), indentation, spacing before and after, borders and shading.
x^2 uses superscript (Ctrl + Shift + P in Writer); H2O uses subscript (Ctrl + Shift + B).
| Feature | Where | Use |
|---|---|---|
| Page size and orientation | Format > Page Style | A4, Letter; portrait / landscape |
| Margins | Format > Page Style > Page | space around text |
| Header and footer | Insert > Header and Footer | text repeated on every page: school name, title |
| Page numbers | Insert > Field > Page Number | numbering pages |
| Page break | Ctrl + Enter | start a new page |
F7; red wavy line = spelling error
fixes common errors as you type (teh -> the)
synonyms (Ctrl + F7)
Tools > Word Count
Title "Science Exhibition" - centre, bold, 20 pt. Body text - justified, 12 pt, 1.5 line spacing. Date - right aligned. Insert a picture with square wrap. Add the school name in the header. Export as PDF and share.
| Format | When to use |
|---|---|
| .odt | default LibreOffice format, editable |
| .docx | to share with MS Word users |
| final copy that looks the same everywhere; not easily edited | |
| .txt | plain text without formatting |
bold, italic, underline
page break
find / find and replace
undo / redo
change case
Find and Replace searches for a word or phrase in a document and replaces it with another, one by one or all at once. Shortcut: Ctrl + H.
Left (Ctrl + L) - text starts at the left margin; Centre (Ctrl + E) - text in the middle, used for headings; Right (Ctrl + R) - text along the right margin, used for dates; Justify (Ctrl + J) - text spread evenly to both margins for a neat look.
An entrepreneur identifies an opportunity, organises resources, takes risks and creates value by starting a business. At Level III we study the values, attitudes and competencies that make an entrepreneur successful.
Values are beliefs that guide our behaviour. Entrepreneurial values guide how a person runs a business.
new products, services and methods
takes own decisions and responsibility
sets high standards of quality
no work is small or big
fair dealings with customers and workers
cares for society and environment
A small dhaba owner in Punjab insists on clean kitchens, fresh food and fair wages, even when costs rise. Customers trust him and his business grows into a chain. Outstanding performance and honesty became his brand.
An attitude is our way of thinking and feeling about something, which shows in behaviour.
| Attitude | Meaning | Example |
|---|---|---|
| Uses imagination and intuition | thinks creatively; trusts gut feeling with facts | designs eco-friendly packaging |
| Takes moderate (calculated) risks | neither reckless nor afraid | tests a product in one area before expanding |
| Enjoys freedom of expression and action | likes to decide and act independently | leaves a job to start own studio |
| Seeks economic opportunities | sees business chances in problems | online tuition during lockdown |
| Believes he/she can change the environment | positive, proactive | starts a waste-recycling startup |
| Analyses situations and plans action | studies the market before acting | market survey before opening a cafe |
| Gets involved in activities | hands-on; learns by doing | works on the shop floor with staff |
Competencies are skills and abilities needed to perform well.
| Competency | Meaning |
|---|---|
| Decisiveness | takes quick, clear decisions |
| Initiative | starts things without waiting |
| Interpersonal skills | gets along with customers, staff, suppliers |
| Perseverance | continues despite failure |
| Organisational skills | plans and arranges resources and time |
| Type | Description | Example |
|---|---|---|
| Innovative | introduces new ideas and technology | an AI-based farming app |
| Imitative | copies successful ideas and adapts them | a local food-delivery app |
| Fabian | cautious; changes only when necessary | a traditional family business |
| Drone | resists change even with losses | a shop refusing digital payments |
| Social | solves a social problem while earning | low-cost solar lamps for villages |
Sources of ideas: personal hobbies and skills, problems faced by people, changing trends (health, online shopping), government schemes, travel and reading, and feedback from customers.
Students waste time searching for second-hand books. Meera starts a WhatsApp group and later an app where seniors sell old books to juniors at half price. It saves money, reuses books (green value) and earns her a small commission.
| Barrier | Solution |
|---|---|
| Lack of money | savings, family, bank loans, Mudra loans, crowdfunding |
| Fear of failure | start small, learn from mistakes |
| Lack of knowledge | training, mentors, online courses |
| Competition | better quality, unique feature, good service |
| Legal formalities | take help from experts; use government portals (Udyam registration) |
In 2008, Deepinder Goyal and Pankaj Chaddah noticed that colleagues wasted time looking for restaurant menus. They scanned menus and put them online as "Foodiebay". The idea grew into Zomato, offering restaurant reviews and later food delivery in hundreds of cities. Values seen: innovativeness (a new solution), outstanding performance (constant improvement) and calculated risk (leaving secure jobs).
Innovativeness, independence, outstanding performance, respect for work, honesty and social responsibility. (Any four.)
Taking calculated risks - testing in one market first; seeking economic opportunities - starting online classes during lockdown; analysing and planning - surveying before opening a cafe; believing one can change the environment - starting a recycling unit. Others: imagination and intuition, freedom of action, getting involved.
A green economy is an economy that is low-carbon, resource-efficient and socially inclusive (UNEP). It aims at economic growth and jobs while reducing pollution and protecting nature. It is the economic path to sustainable development.
safe collection and recycling of old electronics
metro, CNG buses, electric vehicles, cycling
solar, wind, small hydro, biomass
GRIHA / LEED buildings, natural light
rainwater harvesting, drip irrigation, recycling
organic farming, crop rotation
| Sector | Green practice | Green job |
|---|---|---|
| Energy | rooftop solar, wind farms | solar PV installer, wind technician |
| Transport | EVs, charging stations | EV service technician |
| Construction | insulation, solar water heaters | green building auditor |
| Waste | segregation, composting, e-waste recycling | waste-management supervisor |
| Agriculture | organic manure, bio-pesticides | organic farming advisor |
| Water | watershed management | water-quality analyst |
| Initiative | Aim |
|---|---|
| National Action Plan on Climate Change (NAPCC), 2008 | eight national missions on solar energy, energy efficiency, sustainable habitat, water, Himalayan ecosystem, green India, sustainable agriculture, strategic knowledge |
| National Solar Mission | large-scale solar power |
| Swachh Bharat Mission | sanitation and waste management |
| FAME India | faster adoption of electric vehicles |
| UJALA | LED bulbs to save electricity |
| Jal Jeevan Mission / Jal Shakti Abhiyan | tap water for every home; water conservation |
| Mission LiFE | Lifestyle for Environment - mindful use of resources |
| Stakeholder | Role |
|---|---|
| Government | makes laws, gives subsidies (e.g. for solar pumps), sets standards |
| Private companies | adopt clean technology, reduce waste, CSR projects |
| NGOs | awareness, community projects, protecting rights |
| Communities | protect local forests and water bodies |
| Consumers | buy energy-efficient and eco-friendly products |
| Banks | give loans for green projects (green bonds) |
| Research institutions | develop new clean technologies |
green spaces, public transport, waste plants
eco-tourism, homestays, no plastic
cleaner production, effluent treatment
afforestation, community forests
no overfishing, protected breeding seasons
natural farming, solar pumps
Green skills are the knowledge, abilities, values and attitudes needed to live in, develop and support a sustainable and resource-efficient society. Green jobs protect or restore the environment while providing decent work.
Arjun completes a course in solar PV installation under Skill India's Skill Council for Green Jobs (SCGJ). He installs rooftop solar panels for homes under a government subsidy scheme. He earns well, homes save electricity bills, and the country reduces coal use - benefits to him, society and the environment.
| Indicator | What it shows |
|---|---|
| Share of renewable energy | how much power comes from solar, wind, hydro |
| Carbon emissions per person | pollution from energy use |
| Forest cover | protection of land and biodiversity |
| Waste recycled | resource efficiency |
| Number of green jobs | social benefit |
use less electricity
less plastic waste
zero emissions for short trips
metals recovered safely
A green economy is low-carbon, resource-efficient and socially inclusive. Features: uses renewable energy, reduces waste and pollution, creates green jobs.
Government - makes green policies and gives subsidies; industries - adopt clean technology and reduce waste; consumers - buy eco-friendly products; NGOs - spread awareness and run community projects. Banks give green finance and research bodies develop clean technology.
Part A carries 10 marks: four one-mark questions (out of six) and three two-mark questions (out of five). Each unit gives 2 marks.
| Writing skill | Key points | Example |
|---|---|---|
| Sentence vs phrase | sentence = subject + verb, complete sense; phrase = group of words without complete sense | The bell rang. / after the bell |
| Kinds of sentences | declarative, interrogative, imperative, exclamatory | Close the door. / How lovely! |
| Parts of a sentence | subject (who/what) and predicate (verb + object/complement) | Riya (S) wrote a poem (P). |
| Parts of speech | noun, pronoun, verb, adjective, adverb, preposition, conjunction, interjection | She (pron.) quickly (adv.) solved it. |
| Articles | a (consonant sound), an (vowel sound), the (particular) | a university, an hour, the Sun |
| Paragraph | one main idea: topic sentence, supporting sentences, concluding sentence |
| Feature | Use | How / shortcut |
|---|---|---|
| Word processor | create, edit, format, save and print documents | LibreOffice Writer (.odt), MS Word (.docx), Google Docs |
| Open / New / Save / Print | basic file work | Ctrl + O, Ctrl + N, Ctrl + S, Ctrl + P; Save As for other formats; Export as PDF |
| Font | face, size, style, colour | Ctrl + B bold, Ctrl + I italic, Ctrl + U underline |
| Alignment | left, centre, right, justify | Ctrl + L, Ctrl + E, Ctrl + R, Ctrl + J |
| Header and footer | text repeated at the top/bottom of each page; page numbers | Insert > Header and Footer |
| Bullets and numbering | lists | Format > Bullets and Numbering |
| Tables | rows and columns of data | Table > Insert Table (Ctrl + F12) |
| Find and Replace | search and change words | Ctrl + H |
| AutoCorrect | fixes typing errors automatically | Tools > AutoCorrect |
| Text wrapping | how text flows around pictures | right-click picture > Wrap |
| Topic | Key points |
|---|---|
| Green economy | low-carbon, resource-efficient, socially inclusive |
| Main sectors | e-waste management, green transportation (EVs, metro), renewable energy (solar, wind), green construction (GRIHA buildings), water management (rainwater harvesting, drip irrigation) |
| Policy initiatives in India | National Action Plan on Climate Change (National Solar Mission etc.), Swachh Bharat Mission, FAME for EVs, UJALA LED scheme, Jal Jeevan Mission |
| Stakeholders | government, private companies, NGOs, communities, consumers, banks, research bodies |
| Greening sectors | cities, buildings, tourism (eco-tourism), industry, transport, renewable energy, waste, agriculture (organic), water, forests (afforestation), fisheries (no overfishing) |
4 marks · meaning, evolution, types, domains, terminology, benefits and limits
Artificial Intelligence (AI) is the branch of computer science that builds machines and software able to perform tasks that normally need human intelligence - learning, reasoning, problem solving, understanding language, recognising images and taking decisions. John McCarthy, who coined the term in 1956, defined it as "the science and engineering of making intelligent machines".
recognise faces, read number plates
voice assistants, speech-to-text
translate, summarise, chat
improve with more data
recommend, predict, plan routes
robots, self-driving cars
| Year | Event | Why it matters |
|---|---|---|
| 1950 | Alan Turing proposes the Turing Test | can a machine's replies be told apart from a human's? |
| 1956 | Dartmouth Conference | birth of AI as a field |
| 1966 | ELIZA chatbot | early natural language program |
| 1970s-80s | expert systems; AI winters | rule-based systems, then funding drops |
| 1997 | IBM Deep Blue defeats Garry Kasparov at chess | computing power + search |
| 2011 | IBM Watson wins Jeopardy!; Siri launched | question answering, voice assistants |
| 2012 | deep learning wins ImageNet | neural networks become practical |
| 2016 | AlphaGo beats Lee Sedol at Go | reinforcement learning |
| 2022+ | ChatGPT, Gemini, image generators | generative AI for everyone |
Super AI is a hypothetical AI that would be smarter than humans in every field.
| Type | Description | Example |
|---|---|---|
| Reactive machines | react to the present; no memory | Deep Blue chess computer |
| Limited memory | use recent past data to decide | self-driving cars, chatbots |
| Theory of mind | would understand emotions and beliefs | under research |
| Self-aware | would have consciousness | hypothetical |
| Domain | Works on | Applications |
|---|---|---|
| Data science / statistical data | structured data | price prediction, fraud detection, recommendation |
| Computer vision | images, videos | face recognition, medical scans, self-driving cars |
| NLP | human language | translation, chatbots, sentiment analysis |
| Term | Meaning |
|---|---|
| Machine learning (ML) | algorithms that learn patterns from data and improve with experience |
| Deep learning (DL) | ML using artificial neural networks with many layers |
| Neural network | a network of connected nodes (neurons) inspired by the brain |
| Data | facts used to train models |
| Model | the learned function that makes predictions |
| Algorithm | step-by-step procedure to learn or solve |
| Training / testing | learning from data / checking on new data |
| Supervised / unsupervised / reinforcement | learning with labels / without labels / by rewards |
A calculator follows fixed rules - not AI. An e-mail filter that learns which mails are spam from examples - AI (machine learning). A robot vacuum that maps rooms and learns obstacles - AI. A timer-based street light - not AI.
disease detection from scans, drug discovery
crop disease detection, weather-based advice (Kisan apps)
fraud detection, credit scoring, chatbots
navigation, traffic prediction, driver assistance
recommendations, demand forecasting
adaptive learning, automatic grading
Bhashini translation, smart city cameras
games, music and video recommendations
Narrow AI performs a single specific task (e.g. face unlock) and exists today. General AI would perform any intellectual task like a human; it does not exist yet.
AI is the broad field of making machines intelligent. ML is a subset of AI where machines learn from data, e.g. a spam filter. DL is a subset of ML that uses multi-layer neural networks, e.g. face recognition. So DL is inside ML, and ML is inside AI.
Artificial Intelligence is the branch of computer science that builds machines able to do tasks that normally need human intelligence: learning, reasoning, perception, language and decision making. Intelligent machines learn from data and improve with experience.
| Year | Milestone | Why it matters |
|---|---|---|
| 1950 | Alan Turing's paper and the Turing Test | can a machine imitate a human in conversation? |
| 1956 | Dartmouth workshop; John McCarthy coins "AI" | birth of AI as a field |
| 1966 | ELIZA chatbot (MIT) | first conversation program |
| 1970s, late 1980s | AI winters | funding fell when promises were not met |
| 1997 | IBM Deep Blue beats Garry Kasparov | AI wins at chess |
| 2011 | IBM Watson wins Jeopardy! | understanding natural-language questions |
| 2012 | deep learning wins image recognition (ImageNet) | start of the deep learning era |
| 2016 | DeepMind AlphaGo beats Lee Sedol | reinforcement learning in a very complex game |
| 2022 onwards | ChatGPT and other generative AI | AI for everyone |
| Domain | Data | Applications |
|---|---|---|
| Data Science | numbers, tables | price and sales prediction, fraud detection, weather forecasting |
| Computer Vision | images, videos | face recognition, self-driving cars, medical imaging, number-plate reading |
| Natural Language Processing | text, speech | chatbots, translation, sentiment analysis, voice assistants |
| Term | Meaning |
|---|---|
| Machine Learning | computers learn patterns from data without being explicitly programmed |
| Deep Learning | ML using neural networks with many layers |
| Neural network | connected nodes in input, hidden and output layers, inspired by the brain |
| Supervised learning | learns from labelled data (classification, regression) |
| Unsupervised learning | finds patterns in unlabelled data (clustering) |
| Reinforcement learning | learns by rewards and penalties (game AI, robots) |
| Features / label | inputs / output to predict |
| Training / testing data | data to teach / data to check the model |
| Cognitive computing | systems that mimic perception, learning and reasoning (IBM Watson) |
Practical activity: categorise given applications into the three domains (e.g. Google Lens - CV, Alexa - NLP, stock prediction - Data Science).
5 marks · demand, job roles, skills, tools, industries
AI is used in almost every industry, so the demand for people who can build, use and manage AI is rising fast. Reports by NASSCOM and the World Economic Forum list AI and machine learning specialists among the fastest-growing jobs. India's IndiaAI Mission and digital programmes are also creating opportunities.
every app and device creates data
industries want efficiency
startups and companies fund AI
e-governance, UPI, Aadhaar-based services
fewer trained people than jobs
| Job role | What they do | Key skills |
|---|---|---|
| Data scientist | collects, analyses and models data to find insights | statistics, Python, ML |
| Data analyst | cleans data, makes reports and dashboards | Excel, SQL, visualisation |
| Machine learning engineer | builds, trains and deploys ML models | Python, ML libraries, cloud |
| AI research scientist | creates new algorithms | advanced maths, research |
| Computer vision engineer | systems that understand images | OpenCV, CNNs |
| NLP engineer | systems that understand language | text processing, transformers |
| Robotics engineer | designs intelligent robots | electronics, control, AI |
| AI product manager | plans AI products to meet user needs | business + tech understanding |
| AI ethicist / policy expert | ensures fair and responsible AI | ethics, law, social science |
| Prompt engineer | designs prompts for generative AI | language, domain knowledge |
| Data engineer | builds data pipelines and storage | databases, big data tools |
| Skill area | Examples |
|---|---|
| Technical | Python, libraries (NumPy, Pandas, scikit-learn), SQL, cloud basics |
| Mathematical | mean, standard deviation, probability, matrices, graphs |
| Analytical | breaking problems into parts, logical thinking |
| Soft skills | communication, collaboration, curiosity, adaptability |
| Ethical | fairness, privacy, transparency |
| Category | Tools |
|---|---|
| Programming | Python, R, Jupyter Notebook, Google Colab |
| ML libraries | scikit-learn, TensorFlow, PyTorch, Keras |
| Data handling | NumPy, Pandas, SQL, Excel |
| Visualisation | Matplotlib, Seaborn, Tableau, Power BI |
| No-code AI | Teachable Machine, Orange, Lobe |
| Cloud AI | Google Cloud AI, AWS, Microsoft Azure |
| Generative AI | ChatGPT, Gemini, Copilot |
| Industry | AI use | Job example |
|---|---|---|
| Healthcare | reading X-rays, predicting disease | medical AI analyst |
| Finance | fraud detection, robo-advisors | risk-model developer |
| Agriculture | yield prediction, pest detection | agri-data scientist |
| Manufacturing | predictive maintenance, quality check | industrial AI engineer |
| Retail and e-commerce | recommendations, inventory | recommendation-system engineer |
| Education | personalised learning | ed-tech data scientist |
| Media | content recommendation, content generation | content-AI specialist |
A startup uses drone images and AI to detect crop disease early. It needs CV engineers (image models), data scientists (yield prediction), app developers (farmer app in local languages) and agronomists (domain knowledge). One AI product creates many different jobs.
AI changes jobs more than it simply removes them. People who learn to work with AI (upskilling and reskilling) will be in demand.
Data scientist, machine learning engineer, computer vision engineer, NLP engineer, robotics engineer, AI ethicist, AI product manager. (Any four.)
Mathematical skills (statistics, probability, linear algebra); programming in Python and SQL; knowledge of machine-learning concepts and tools; data handling and visualisation; domain knowledge; and soft skills such as communication, teamwork, problem solving and ethics.
AI is used in almost every industry, so the global demand for AI professionals is rising faster than the supply of skilled people. AI careers are open to students from science, commerce and humanities who build the right skills.
| Job role | What they do | Key skills |
|---|---|---|
| Data Scientist | collects, cleans and analyses data; builds predictive models | Python, statistics, ML, communication |
| Machine Learning Engineer | builds, trains and deploys ML models in products | Python, ML frameworks, software engineering, cloud |
| Data Analyst | reports and dashboards for business decisions | Excel, SQL, Pandas, Tableau/Power BI |
| AI Research Scientist | invents new algorithms; publishes research | advanced maths, deep learning |
| NLP Engineer | chatbots, translation, speech systems | linguistics, NLP libraries |
| Computer Vision Engineer | image and video AI | OpenCV, CNNs |
| Robotics Engineer | robots that sense and act | electronics, control, AI |
| AI Ethicist | checks fairness, privacy and accountability | ethics, law, communication |
| AI Product Manager | decides what the AI product should do | business sense, communication |
| Data Annotator | labels data for training | attention to detail |
| Industry | Uses of AI |
|---|---|
| Healthcare | X-ray and scan reading, disease prediction, drug discovery, symptom chatbots |
| Finance | fraud detection, credit scoring, trading, customer chatbots |
| Agriculture | crop disease detection, yield and weather prediction, smart irrigation, drones |
| Retail and e-commerce | recommendations, demand forecasting, dynamic pricing |
| Manufacturing | predictive maintenance, quality inspection, robots |
| Transport | self-driving cars, traffic prediction, route optimisation |
| Education | personalised learning, automatic grading, tutoring bots |
Practical: list ten companies hiring for AI roles (LinkedIn, Naukri, company sites) and compare the technical and soft skills asked by two companies for one role.
5 marks · tokens, data types, operators, control statements, CSV, NumPy, Pandas, scikit-learn
Python is a free, open-source, high-level, interpreted language created by Guido van Rossum (1991). It is the most popular language for AI because of its simple syntax and powerful libraries (NumPy, Pandas, scikit-learn, TensorFlow).
English-like syntax
NumPy, Pandas, Matplotlib, scikit-learn
many tutorials and answers
test line by line
Windows, Linux, macOS
The smallest units of a Python program are tokens.
| Token | Meaning | Examples |
|---|---|---|
| Keywords | reserved words | if, else, for, while, def, True, None |
| Identifiers | names given by the programmer | total, marks_1, Student |
| Literals | fixed values | 25, 3.14, "AI", True |
| Operators | perform operations | + - * / // % ** == and |
| Punctuators | symbols for structure | ( ) [ ] { } : , # |
x = 25; y = 3.5; name = "Kavya"; ok = True
marks = [78, 85, 90] # list - mutable
point = (3, 4) # tuple - immutable
student = {"name": "Kavya", "class": 11} # dictionary
print(type(x), type(y), type(name), type(ok))
print(marks[0], point[1], student["name"])
| Type | Operators | Example (a = 17, b = 5) |
|---|---|---|
| Arithmetic | + - * / // % ** | a // b = 3, a % b = 2, b ** 2 = 25 |
| Relational | == != > < >= <= | a > b gives True |
| Logical | and, or, not | a > 10 and b > 10 gives False |
| Assignment | = += -= *= | a += 1 makes a 18 |
| Membership | in, not in | 3 in [1, 2, 3] gives True |
marks = 72
if marks >= 90:
grade = "A1"
elif marks >= 75:
grade = "A2"
elif marks >= 33:
grade = "Pass"
else:
grade = "Fail"
print("Grade:", grade)
total = 0
for n in range(1, 11):
total += n
print("Sum of 1 to 10 =", total)
n = 5; fact = 1
while n > 1:
fact *= n
n -= 1
print("5! =", fact)
break stops the loop; continue skips to the next round.
L = [12, 45, 7, 30]
L.append(50); L.sort()
print(L, len(L), max(L), sum(L) / len(L))
print(L[1:3], L[-1])
d = {"Amit": 82, "Sara": 91}
d["Ravi"] = 76
for k, v in d.items():
print(k, "->", v)
| Method | Use |
|---|---|
| append(x) | add at the end |
| insert(i, x) | add at index i |
| remove(x) / pop(i) | delete by value / by index |
| sort() / reverse() | arrange / reverse |
| keys(), values(), items() | dictionary parts |
def area_circle(r):
return 3.14159 * r * r
def grade(m):
return "Pass" if m >= 33 else "Fail"
print(round(area_circle(7), 2), grade(45))
A CSV (Comma Separated Values) file stores table data as plain text, one row per line. AI datasets are often CSV files.
import csv
with open("marks.csv", "w", newline="") as f: # write a CSV file
w = csv.writer(f)
w.writerow(["name", "marks"])
w.writerows([["Asha", 88], ["Ben", 67], ["Chitra", 92]])
with open("marks.csv") as f: # read it back
r = csv.reader(f)
header = next(r)
for row in r:
print(row[0], "->", int(row[1]) + 5) # 5 grace marks
NumPy gives fast arrays and maths; Pandas gives the DataFrame - a table with rows and columns.
import numpy as np a = np.array([10, 20, 30, 40]) print(a.mean(), a.max(), a * 2) # 25.0 40 [20 40 60 80] m = np.array([[1, 2], [3, 4]]) print(m.shape) print(m.T) # transpose
import pandas as pd
import numpy as np
data = {"name": ["Asha", "Ben", "Chitra", "Dev", "Esha"], "marks": [78, np.nan, 92, 60, 85]}
pd.DataFrame(data).to_csv("students.csv", index=False) # make the CSV file
df = pd.read_csv("students.csv")
print(df.head()) # first 5 rows
print(df.describe()) # count, mean, std, min, max
df["marks"] = df["marks"].fillna(df["marks"].mean()) # fill the missing value
print(df[df["marks"] > 80]) # filter rows
A list is mutable (its items can be changed) and uses square brackets [ ]. A tuple is immutable and uses round brackets ( ). Tuples are faster and safe for fixed data.
Read a, b, c. If a >= b and a >= c, print a; elif b >= c, print b; else print c. (Or use max(a, b, c).)
Python (Guido van Rossum, 1991) is a free, interpreted, high-level language - the most used language in AI. Programs below have a Run button: open them in the built-in Python and change them.
| Topic | Key points |
|---|---|
| Character set | letters A-Z a-z, digits 0-9, special symbols, whitespace, Unicode |
| Tokens | keywords (if, for, True), identifiers (names), literals (25, 3.5, "AI", True, None), operators (+, ==, and), punctuators ( ) [ ] { } , : . |
| Modes | interactive (>>> shell, one statement) and script (.py file) |
| Data types | int, float, complex, str, bool, list (mutable), tuple (immutable), dict (key:value), set (unique), None |
| Operators | arithmetic + - * / // % **; relational == != > < >= <=; logical and or not; assignment = += -=; membership in, not in; identity is, is not |
| Control statements | if, if-else, if-elif-else; for, while; break, continue, pass; loop-else |
# data types and type()
values = [25, 3.5, 2 + 3j, "AI", True, [1, 2], (1, 2), {"a": 1}, {1, 2}, None]
for v in values:
print(repr(v), "->", type(v).__name__)Output:
25 -> int
3.5 -> float
(2+3j) -> complex
'AI' -> str
True -> bool
[1, 2] -> list
(1, 2) -> tuple
{'a': 1} -> dict
{1, 2} -> set
None -> NoneType# operators and precedence: ** then * / // % then + - then comparisons then not, and, or
print(2 + 3 * 4 ** 2, (2 + 3) * 4, 17 // 5, 17 % 5)
print(5 > 3 and 2 > 4, not 5 > 3 or 2 < 4)
print("a" in "data", [1, 2] == [1, 2])Output:
50 20 3 2 False True True True
# selection and iteration
marks = [92, 45, 76, 30, 64]
for m in marks:
if m >= 90:
g = "A"
elif m >= 60:
g = "B"
elif m >= 33:
g = "C"
else:
g = "Fail"
print(m, g)
n = 5
f = 1
while n > 1:
f *= n
n -= 1
print("5! =", f)Output:
92 A 45 C 76 B 30 Fail 64 B 5! = 120
s = "Artificial Intelligence"
print(len(s), s[0], s[-1], s[0:10], s.upper(), s.split())
L = [10, 20, 30]
L.append(40); L.insert(0, 5); L.remove(20)
print(L, L[1:3], sum(L), max(L))
t = (3, 1, 2)
print(sorted(t), t.count(1))
d = {"Asha": 90, "Ravi": 75}
d["Meena"] = 88
print(d.keys(), d.get("Ravi"), len(d))Output:
23 A e Artificial ARTIFICIAL INTELLIGENCE ['Artificial', 'Intelligence'] [5, 10, 30, 40] [10, 30] 85 40 [1, 2, 3] 1 dict_keys(['Asha', 'Ravi', 'Meena']) 75 3
import csv
with open("rain.csv", "w", newline="") as f:
w = csv.writer(f)
w.writerow(["Month", "Rain"])
w.writerows([["Jan", 20], ["Feb", 35], ["Mar", 60]])
with open("rain.csv") as f:
for row in csv.reader(f):
print(row)Output:
['Month', 'Rain'] ['Jan', '20'] ['Feb', '35'] ['Mar', '60']
import numpy as np a = np.array([[1, 2, 3], [4, 5, 6]]) print(a.shape, a.ndim, a.size) print(a * 10) print(a.mean(), a.max(), a.sum(axis=0)) print(np.arange(0, 10, 2), np.zeros(3), np.ones((2, 2)))
Output:
(2, 3) 2 6 [[10 20 30] [40 50 60]] 3.5 6 [5 7 9] [0 2 4 6 8] [0. 0. 0.] [[1. 1.] [1. 1.]]
import pandas as pd
df = pd.DataFrame({"Month": ["Jan", "Feb", "Mar", "Apr"], "Rain": [20, 35, 60, 120]})
print(df)
print(df.head(2))
print(df["Rain"].mean(), df["Rain"].max())
print(df[df["Rain"] > 30])Output:
Month Rain 0 Jan 20 1 Feb 35 2 Mar 60 3 Apr 120 Month Rain 0 Jan 20 1 Feb 35 58.75 120 Month Rain 1 Feb 35 2 Mar 60 3 Apr 120
Scikit-learn (sklearn) gives ready ML algorithms. The pattern is always: import - create model - fit - predict - evaluate. (Run this in Anaconda / Jupyter; it is not part of the built-in browser Python.)
from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
import pandas as pd
# a small rainfall dataset (in Jupyter you would use pd.read_csv("rainfall.csv"))
df = pd.DataFrame({"Temperature": [20, 22, 25, 27, 30, 32, 35, 37, 40, 42],
"Rainfall": [120, 115, 100, 95, 80, 72, 60, 55, 40, 35]})
X = df[["Temperature"]] # feature(s)
y = df["Rainfall"] # label
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1)
model = LinearRegression()
model.fit(X_train, y_train) # train
print("slope:", model.coef_, "intercept:", round(model.intercept_, 2))
print("predictions:", model.predict(X_test)) # predict
print("R2 score:", round(model.score(X_test, y_test), 3))int(input()) for whole numbers.a * 2 doubles every element.= assigns, == compares.range(1, 5) stops at 4.class, for cannot be variable names.5 marks · design thinking, empathy map, 5W1H, SDGs, capstone stages
A capstone project is a final project in which students apply everything they have learnt to solve a real-world problem. In AI, it means identifying a problem, collecting data, building an AI model or prototype, testing it and presenting the solution. It develops problem solving, teamwork, research and communication.
Design thinking is a human-centred, creative problem-solving approach. It starts with understanding the people who face the problem.
| Stage | Purpose | Activity | Example: students skipping breakfast |
|---|---|---|---|
| Empathise | understand users' feelings and needs | interviews, observation, empathy map | talk to students and parents |
| Define | state the core problem clearly | problem statement, 5W1H | "Students in the morning shift skip breakfast because they leave early." |
| Ideate | generate many solutions | brainstorming, mind maps | reminder app, quick-recipe chatbot, canteen pre-orders |
| Prototype | build a simple model | paper sketch, app mock-up, basic model | chatbot prototype |
| Test | try with users and improve | feedback, observation | students try the chatbot; improve it |
after testing, teams often go back to define or ideate again.
An empathy map is a tool to understand the user deeply. It has four quadrants:
what the user says aloud: "I have no time in the morning."
what the user thinks: "Breakfast is not important."
actions: skips breakfast, buys chips at break
emotions: tired, hungry, irritated
Some maps also include pains (frustrations) and gains (hopes).
Who: farmers in a village. What: crops are damaged by pests found too late. Where: cotton fields. When: monsoon season. Why: losses reduce income. How: an app that detects pests from leaf photos. Problem statement: "Cotton farmers lose income during the monsoon because pests are detected late; an AI app can identify pests early from leaf photos."
The 17 Sustainable Development Goals give meaningful problem areas for AI projects.
| SDG | Project idea |
|---|---|
| 2 Zero hunger | crop-disease detection |
| 3 Good health | early disease prediction from symptoms |
| 4 Quality education | personalised quiz app |
| 6 Clean water | water-quality prediction |
| 11 Sustainable cities | traffic or waste-bin monitoring |
| 13 Climate action | weather / flood prediction |
| Stage | Key questions | Output |
|---|---|---|
| Problem identification | Is it real? Can AI help? | problem statement |
| Data collection | What data? From where? Is it ethical? | dataset |
| Exploration and preparation | missing values? patterns? | clean data, graphs |
| Modelling | classification, regression or clustering? | trained model |
| Evaluation | how accurate is it? | metrics, confusion matrix |
| Deployment | how will users use it? | app, website, report |
Problem: manual attendance wastes 10 minutes per class. Empathise: teachers and students interviewed. Data: 50 photos of each student (with consent). Model: face recognition using a pre-trained model. Evaluation: 95% accuracy in good light, lower in poor light. Ethics: consent, secure storage, delete data after the project.
Empathise, Define, Ideate, Prototype and Test.
An empathy map is a tool used in the empathise stage to understand the user. Says - what the user says; Thinks - what the user thinks but may not say; Does - the user's actions and behaviour; Feels - the user's emotions and worries.
Design thinking is a human-centred, iterative way of solving problems creatively. It is used to plan the capstone project.
| Stage | What we do | Tools |
|---|---|---|
| Empathize | understand users: observe, interview, experience their problem | interviews, observation, empathy map |
| Define | write a clear, user-centred problem statement | point-of-view statement, "How might we...?" |
| Ideate | generate many ideas without judging | brainstorming, mind maps, SCAMPER, sticky notes |
| Prototype | build a quick, cheap model of the best idea | sketches, paper models, mock-up screens, no-code AI |
| Test | try with users, collect feedback, improve | user testing, feedback forms |
| Question | Asks | Example: plastic waste in school |
|---|---|---|
| Who | who is affected? | students, canteen staff, cleaners |
| What | what is the problem? | plastic wrappers and bottles everywhere |
| When | when does it happen? | after lunch break, events |
| Where | where does it happen? | canteen, playground |
| Why | why does it matter? | pollution, blocked drains (SDG 12, 14) |
| How | how can it be solved? | AI bin that sorts waste, awareness, refill stations |
17 goals adopted by the United Nations in 2015, to be achieved by 2030. A capstone project must be aligned to at least one SDG.
| SDG | Goal | AI idea |
|---|---|---|
| 1 | No Poverty | job-matching app |
| 2 | Zero Hunger | crop disease detector |
| 3 | Good Health and Well-being | early disease detection |
| 4 | Quality Education | personal tutor bot |
| 6 | Clean Water and Sanitation | water quality prediction |
| 7 | Affordable and Clean Energy | solar output forecast |
| 11 | Sustainable Cities | smart traffic signals |
| 12 | Responsible Consumption | food-waste prediction |
| 13 | Climate Action | flood and heatwave alerts |
Each stage is documented as in the CBSE IBM Projects Cookbook. A project abstract gives the title, problem, users, SDG, solution, data needed and expected impact in simple, non-technical words.
Practical: create an empathy map for a given scenario, and a project abstract using the design thinking framework.
6 marks · collection, levels of measurement, statistics, charts, matrices, pre-processing
Data literacy is the ability to read, understand, analyse, create and communicate with data. AI runs on data, so an AI student must know how data is collected, measured, summarised, cleaned and shown.
| Method | Example |
|---|---|
| Survey / questionnaire | Google Form on students' screen time |
| Interview | talking to farmers about crop loss |
| Observation | counting vehicles at a crossing |
| Experiment | testing plant growth with fertilisers |
| Sensors / IoT | temperature and humidity logs |
| Web scraping / APIs | prices from websites (with permission) |
| Level | Properties | Examples | Allowed statistics |
|---|---|---|---|
| Nominal | categories, no order | blood group, gender, colour | mode, counts |
| Ordinal | categories with order | ranks, grades A/B/C, rating 1-5 | median, mode |
| Interval | ordered, equal gaps, no true zero | temperature in degree C, year | mean, difference |
| Ratio | equal gaps and true zero | height, weight, marks, income | all: mean, ratio ("twice as heavy") |
20 degree C is not "twice as hot" as 10 degree C because 0 degree C is not "no heat" - temperature in Celsius is interval, not ratio.
| Measure | Formula / rule |
|---|---|
| Mean | sum of values / number of values |
| Median | middle value of sorted data (average of two middle values if n is even) |
| Mode | most frequent value |
| Measure | Meaning |
|---|---|
| Range | maximum - minimum |
| Variance | average of squared deviations from the mean |
| Standard deviation (SD) | square root of variance; typical distance from the mean |
Data: 2, 4, 4, 4, 5, 5, 7, 9 (n = 8). Mean = 40 / 8 = 5. Deviations squared: 9, 1, 1, 1, 0, 0, 4, 16; sum = 32. Variance = 32 / 8 = 4. SD = square root of 4 = 2. Median = (4 + 5) / 2 = 4.5; Mode = 4; Range = 9 - 2 = 7.
data = [2, 4, 4, 4, 5, 5, 7, 9]
n = len(data); mean = sum(data) / n
var = sum((x - mean) ** 2 for x in data) / n
print("mean", mean, "variance", var, "SD", var ** 0.5)
compare categories
trends over time
parts of a whole (%)
relation between two variables
distribution of continuous data in ranges
median, quartiles and outliers
A matrix is a rectangular arrangement of numbers in rows and columns. Images, datasets and neural-network weights are stored as matrices.
| Operation | Rule | Example |
|---|---|---|
| Order | rows x columns | [[1, 2, 3], [4, 5, 6]] is 2 x 3 |
| Addition | add matching elements (same order) | [[1, 2]] + [[3, 4]] = [[4, 6]] |
| Scalar multiplication | multiply every element | 2 x [[1, 3]] = [[2, 6]] |
| Transpose | rows become columns | [[1, 2], [3, 4]] -> [[1, 3], [2, 4]] |
| Multiplication | row x column, sum of products | (m x n) x (n x p) gives (m x p) |
A = [[1, 2], [3, 4]], B = [[5, 6], [7, 8]]. AB = [[1x5 + 2x7, 1x6 + 2x8], [3x5 + 4x7, 3x6 + 4x8]] = [[19, 22], [43, 50]].
| Problem | Solution |
|---|---|
| Missing values | remove the row, or fill with mean / median / mode |
| Duplicates | delete repeated rows |
| Outliers | check; correct errors, cap or remove |
| Text categories | encode as numbers (e.g. Male = 0, Female = 1) |
| Different scales | normalise: (x - min) / (max - min) gives 0 to 1 |
Marks 40, 60, 100 (min 40, max 100). Normalised: (40 - 40)/60 = 0, (60 - 40)/60 = 0.33, (100 - 40)/60 = 1.
Both have order and equal gaps, but ratio data has a true zero (e.g. weight), so ratios are meaningful. Interval data has no true zero (e.g. temperature in degree C).
Mean = 45 / 5 = 9; median = 8; mode = 8; range = 13 - 6 = 7.
Data literacy is the ability to read, understand, analyse and communicate with data. AI is only as good as its data.
Collected first-hand for the purpose: surveys, interviews, observation, experiments, sensors.
Collected earlier by others: census, research papers, data.gov.in, Kaggle, WHO and World Bank data.
| Level | Property | Examples | Allowed statistics |
|---|---|---|---|
| Nominal | names / categories only | gender, blood group, city | mode, counts |
| Ordinal | order, unequal gaps | ranks, grades, ratings | median, mode |
| Interval | equal gaps, no true zero | temperature in deg C, years | mean, SD, + and - |
| Ratio | equal gaps and a true zero | height, weight, income, rainfall | all, including ratios |
| Measure | Formula | Data 2, 4, 4, 4, 5, 5, 7, 9 |
|---|---|---|
| Mean | sum / n | 40 / 8 = 5 |
| Median | middle of sorted data | (4 + 5) / 2 = 4.5 |
| Mode | most frequent | 4 |
| Range | max - min | 9 - 2 = 7 |
| Variance | sum of (x - mean)^2 / n | (9+1+1+1+0+0+4+16) / 8 = 32 / 8 = 4 |
| Standard deviation | square root of variance | sqrt(4) = 2 |
import statistics as st
data = [2, 4, 4, 4, 5, 5, 7, 9]
print("Mean :", st.mean(data))
print("Median :", st.median(data))
print("Mode :", st.mode(data))
print("Variance:", st.pvariance(data), " (sample:", round(st.variance(data), 3), ")")
print("SD :", st.pstdev(data))Output:
Mean : 5 Median : 4.5 Mode : 4 Variance: 4 (sample: 4.571 ) SD : 2.0
| Chart | Best for | matplotlib |
|---|---|---|
| Line graph | trend over time (monthly rainfall) | plt.plot(x, y) |
| Bar graph | compare categories | plt.bar(x, y) |
| Histogram | distribution of numeric values in bins | plt.hist(data, bins=5) |
| Scatter plot | relation between two variables | plt.scatter(x, y) |
| Pie chart | parts of a whole | plt.pie(values, labels=...) |
import matplotlib.pyplot as plt
months = ["Jan", "Feb", "Mar", "Apr", "May", "Jun"]
rain = [20, 35, 60, 120, 210, 95]
plt.plot(months, rain, marker="o")
plt.title("Monthly rainfall (mm)")
plt.xlabel("Month")
plt.ylabel("Rainfall")
plt.show()
| Idea | Meaning | Example |
|---|---|---|
| Matrix, order | numbers in rows and columns; order = rows x columns | [1 2 3; 4 5 6] is 2 x 3 |
| Types | row, column, square, zero, diagonal, identity (1s on diagonal) | I = [1 0; 0 1] |
| Addition | same order; add corresponding elements | [1 2; 3 4] + [5 6; 7 8] = [6 8; 10 12] |
| Scalar multiple | multiply every element | 2[1 2; 3 4] = [2 4; 6 8] |
| Multiplication | columns of A = rows of B; row x column | [1 2; 3 4] x [5 6; 7 8] = [19 22; 43 50] |
| Transpose | rows become columns | [1 2; 3 4]T = [1 3; 2 4] |
| Uses in AI | images (pixel matrices), datasets (rows = records), neural network weights, recommendations |
| Step | What is done |
|---|---|
| Cleaning | handle missing values (fill with mean/median or remove), remove duplicates, fix wrong entries and outliers |
| Transformation | normalisation (x - min)/(max - min), standardisation, encoding categories into numbers |
| Reduction | remove features that do not help; sample large data |
| Integration | combine data from many sources |
| Train-test split | e.g. 80% training, 20% testing |
| Evaluation | check the model on test data: accuracy, error; improve if needed |
statistics.variance divides by n - 1 (sample). For the population formula use pvariance or numpy.var.6 marks · types of ML, correlation, linear regression, classification and k-NN, clustering and k-means
Machine learning (ML) is a subset of AI in which computers learn patterns from data and improve their performance without being explicitly programmed with every rule (Arthur Samuel, 1959).
| Type | Data | Goal | Examples |
|---|---|---|---|
| Supervised | labelled (input + correct output) | predict output for new input | spam detection, house prices |
| Unsupervised | unlabelled | find hidden groups / patterns | customer segmentation |
| Reinforcement | feedback as rewards | learn the best actions | game-playing AI, robots |
Correlation measures how strongly two variables are related. The Pearson correlation coefficient (r) ranges from -1 to +1.
| r | Meaning | Example |
|---|---|---|
| +1 / close to +1 | strong positive: both rise together | study hours and marks |
| 0 | no linear relation | shoe size and IQ |
| -1 / close to -1 | strong negative: one rises, other falls | speed and travel time |
Correlation is not causation: ice-cream sales and drowning cases both rise in summer, but one does not cause the other - heat causes both.
Linear regression is a supervised algorithm that predicts a continuous number by fitting a straight line y = mx + c (m = slope, c = intercept) through the data, using the least squares method.
Study hours x: 1, 2, 3, 4, 5; marks y: 50, 55, 65, 70, 80. xbar = 3, ybar = 64. Deviations x: -2, -1, 0, 1, 2; y: -14, -9, 1, 6, 16. Sum of products = 28 + 9 + 0 + 6 + 32 = 75; sum of x-deviations squared = 10. m = 75 / 10 = 7.5; c = 64 - 7.5 x 3 = 41.5. Line: y = 7.5x + 41.5. For 6 hours: y = 45 + 41.5 = 86.5 marks.
x = [1, 2, 3, 4, 5]; y = [50, 55, 65, 70, 80]
xb = sum(x) / 5; yb = sum(y) / 5
m = sum((a - xb) * (b - yb) for a, b in zip(x, y)) / sum((a - xb) ** 2 for a in x)
c = yb - m * xb
print("m =", m, "c =", c, "prediction for 6 h =", m * 6 + c)
Errors: MAE (mean absolute error), MSE (mean squared error), RMSE (square root of MSE) measure how far predictions are from actual values.
Classification predicts a category (spam / not spam). KNN classifies a new point by looking at its K nearest labelled points and taking the majority vote.
Fruit data (weight g, sweetness): A (150, 7) apple, B (160, 8) apple, C (120, 3) lemon, D (110, 2) lemon, E (155, 6) apple. New fruit (140, 6). Distances: A = sqrt(100 + 1) = 10.05; B = sqrt(400 + 4) = 20.1; C = sqrt(400 + 9) = 20.2; D = sqrt(900 + 16) = 30.3; E = sqrt(225 + 0) = 15. Nearest 3: A, E, B - all apples. Prediction: apple.
Clustering groups similar unlabelled items. K-means forms K clusters:
Example: an online shop groups customers into "budget buyers", "regular buyers" and "premium buyers" from spending data, without any labels.
| Algorithm | Type | Output | Use case |
|---|---|---|---|
| Linear regression | supervised | number | price, marks, sales prediction |
| KNN | supervised | class | disease yes/no, fruit type |
| Decision tree | supervised | class / number | loan approval |
| K-means | unsupervised | groups | customer segmentation |
Regression predicts a continuous numeric value (e.g. house price). Classification predicts a category or class (e.g. spam or not spam).
Choose K and place K initial centroids; assign each data point to the nearest centroid; recompute each centroid as the mean of its points; repeat assignment and update until the centroids no longer change. The final groups are the clusters.
Machine learning lets computers learn patterns from data and improve with experience instead of following only fixed rules.
| Type | Data | Tasks | Example |
|---|---|---|---|
| Supervised | labelled | regression (number), classification (class) | house price; spam filter |
| Unsupervised | unlabelled | clustering, association | customer groups; items bought together |
| Reinforcement | rewards / penalties | sequential decisions | game AI, robot walking |
Correlation measures how strongly two variables move together. Pearson's r lies between -1 and +1: +1 perfect positive, -1 perfect negative, 0 no linear relation. Correlation is not causation.
r = [ n.Sxy - Sx.Sy ] / sqrt( [ n.Sx^2 - (Sx)^2 ] x [ n.Sy^2 - (Sy)^2 ] ) In MS Excel: =CORREL(A2:A8, B2:B8) or =PEARSON(A2:A8, B2:B8)
Linear regression fits the straight line y = mx + c that best describes the data. The best-fit line minimises the sum of squared residuals (actual - predicted): the least squares method.
m (slope) = [ n.Sxy - Sx.Sy ] / [ n.Sx^2 - (Sx)^2 ]
c (intercept) = mean(y) - m x mean(x)
Prediction : y = m x (new x) + c
In MS Excel: scatter chart > Add Trendline > Linear > Display equation on chart
or =SLOPE(y_range, x_range) and =INTERCEPT(y_range, x_range)| Type of classification | Example |
|---|---|
| Binary | spam / not spam, pass / fail |
| Multi-class | digit 0-9, type of fruit |
| Multi-label | a movie tagged comedy and romance |
| Imbalanced | fraud detection (very few fraud cases) |
| Step | k-Nearest Neighbours |
|---|---|
| 1 | choose k (odd for two classes, e.g. 3 or 5) |
| 2 | find the distance of the new point from every training point: d = sqrt((x1 - x2)^2 + (y1 - y2)^2) |
| 3 | select the k nearest points |
| 4 | majority vote of their classes gives the class of the new point |
| Type of clustering | Idea |
|---|---|
| Partitioning (k-means) | split data into k groups around centroids |
| Hierarchical | build a tree (dendrogram) by merging or splitting clusters |
| Density-based (DBSCAN) | clusters are dense regions; sparse points are noise |
| Step | k-means |
|---|---|
| 1 | choose k and k starting centroids |
| 2 | assign each point to the nearest centroid |
| 3 | recompute each centroid as the mean of its points |
| 4 | repeat 2 and 3 until the centroids do not change |
# scikit-learn versions (same code works in Jupyter)
from sklearn.neighbors import KNeighborsClassifier
from sklearn.cluster import KMeans
X_train = [[1, 1], [1, 2], [2, 2], [6, 6], [7, 6], [6, 7]]
y_train = ["A", "A", "A", "B", "B", "B"]
knn = KNeighborsClassifier(n_neighbors=3).fit(X_train, y_train)
print("k-NN prediction for (3, 3):", knn.predict([[3, 3]]))
km = KMeans(n_clusters=2, n_init=10).fit(X_train)
print("cluster labels:", km.labels_)
print("centroids:")
print(km.cluster_centers_)5 marks · language complexity, NLP tasks, phases of NLP, chatbots, applications
Human language is rich, creative and ambiguous. Linguistics is the scientific study of language - its sounds, words, grammar and meaning. Natural Language Processing (NLP) combines linguistics and computer science so that computers can understand, interpret and generate human language.
"I saw a man with a telescope" - who has it?
22 scheduled languages in India alone
"Great, another Monday!"
"bank" of a river vs a money bank
spelling errors, slang, emojis
| Level | Studies | Example |
|---|---|---|
| Phonology | sounds | "cat" = k + a + t |
| Morphology | word structure | un + happy + ness |
| Syntax | grammar and sentence structure | subject - verb - object |
| Semantics | meaning | "bright" = shining or intelligent |
| Pragmatics | meaning in context | "Can you pass the salt?" is a request |
| Discourse | meaning across sentences | "Ravi came. He sat." (He = Ravi) |
| Phase | What it does | Example |
|---|---|---|
| Lexical (morphological) | breaks text into words/tokens and analyses word structure | "Students are reading" -> Students / are / reading |
| Syntactic (parsing) | checks grammar and builds a parse tree | "Reading students are" is rejected |
| Semantic | finds literal meaning; rejects meaningless sentences | "Hot ice cream" flagged as odd |
| Discourse integration | meaning depends on earlier sentences | "Asha lost her pen. She found it later." |
| Pragmatic | real intended meaning in the situation | "It's cold here" may mean "close the window" |
| Step | Example |
|---|---|
| Tokenisation | "AI is fun!" -> AI / is / fun / ! |
| Stop-word removal | remove is, the, a, an, of |
| Stemming | playing, played -> play; studies -> studi |
| Lemmatisation | studies -> study; better -> good |
| POS tagging | AI (noun) is (verb) fun (adjective) |
| Named Entity Recognition | "Sundar Pichai (PERSON) leads Google (ORG) in California (PLACE)" |
Doc 1: "I love AI". Doc 2: "I love coding". Vocabulary: I, love, AI, coding. Vectors: Doc 1 = [1, 1, 1, 0]; Doc 2 = [1, 1, 0, 1].
A word present in every document gets IDF = log(1) = 0 - it is not useful for telling documents apart. Rare, focused words get high TF-IDF.
Modern NLP represents words as vectors so that similar words are close: king - man + woman is close to queen. Large language models use such embeddings and transformers.
pos = {"good", "great", "love", "excellent"}
neg = {"bad", "poor", "hate", "boring"}
review = "the movie was great and the songs were excellent but the story was boring"
words = review.split()
score = sum(w in pos for w in words) - sum(w in neg for w in words)
print("score", score, "->", "positive" if score > 0 else "negative" if score < 0 else "neutral")
Bhashini (National Language Translation Mission) builds translation and speech tools for Indian languages; IndicTrans and AI4Bharat create open models. Challenges include many scripts, code-mixing (Hinglish) and less digital data.
Stemming cuts word endings and may give a non-word (studies -> studi). Lemmatisation uses a dictionary to give the proper base word (studies -> study).
Lexical analysis splits text into words and studies word structure; syntactic analysis checks grammar; semantic analysis finds the literal meaning; discourse integration uses earlier sentences for meaning; pragmatic analysis finds the intended meaning in the real situation.
Natural Language Processing (NLP) combines linguistics and computer science so that computers can understand, interpret and generate human language.
| Challenge | Example |
|---|---|
| Lexical ambiguity (word) | "bank" of a river vs a money bank |
| Syntactic ambiguity (structure) | "I saw the man with the telescope" |
| Context and reference | "He said it was fine" - who is he, what is it? |
| Sarcasm and idioms | "Great, another Monday!"; "break a leg" |
| Variety | spelling errors, slang, accents, code-mixing (Hinglish) |
| Phase | Work | Example |
|---|---|---|
| Lexical analysis | split text into sentences, words (tokens); identify word forms | "AI is fun." -> AI | is | fun | . |
| Syntactic analysis | check grammar and build a parse tree | "Fun is AI the" is rejected |
| Semantic analysis | check the literal meaning | "hot ice-cream" is rejected |
| Discourse integration | use previous sentences | "Riya bought a phone. It is red." - it = phone |
| Pragmatic analysis | intended meaning from context and world knowledge | "Can you pass the salt?" = a request |
| Task | Meaning | Example |
|---|---|---|
| Sentiment analysis | positive / negative / neutral opinion | product reviews, tweets |
| Emotion detection | joy, anger, sadness, fear | customer support priority |
| Text classification | assign categories | spam filter, news topics |
| Chatbots | conversation with users | bank help bot, ChatGPT |
| Machine translation | one language to another | Google Translate |
| Speech recognition / synthesis | speech to text / text to speech | voice typing, screen readers |
| Summarisation, autocomplete | short summary, next word | news apps, Gmail |
Fixed rules and keywords; easy to build; fails on unexpected questions.
Menu bot on a websiteUses NLP and ML; understands varied language; improves with data.
ChatGPT, Gemini, Alexa| Term | Meaning | Ice-cream bot |
|---|---|---|
| Agent | the chatbot | IceCreamBot |
| Intent | what the user wants | OrderIceCream, OrderStatus |
| Training phrases | examples of user messages | "I want two chocolate cones" |
| Entities | details to extract | flavour, size, quantity |
| Responses | replies, using the parameters | "2 chocolate cones ordered!" |
| Fulfilment | connect to a program or database | save the order |
# a simple rule-based chatbot in Python
rules = {"hello": "Hi! Welcome to Scoops.", "flavour": "We have chocolate, vanilla and mango.", "price": "Each scoop costs Rs 50.", "bye": "Thank you, visit again!"}
for msg in ["Hello", "Which flavour do you have?", "What is the price?", "bye"]:
reply = "Sorry, please ask about flavour or price."
for key in rules:
if key in msg.lower():
reply = rules[key]
print("User:", msg, "| Bot:", reply)Output:
User: Hello | Bot: Hi! Welcome to Scoops. User: Which flavour do you have? | Bot: We have chocolate, vanilla and mango. User: What is the price? | Bot: Each scoop costs Rs 50. User: bye | Bot: Thank you, visit again!
IBM Project Debater (2019) was the first AI to debate a human champion live; it searched about 400 million articles to build arguments and rebuttals. Practical: write an article on its interesting facts; build an ice-cream ordering chatbot (Dialogflow / Botsify / Botpress).
4 marks · five pillars, bias and its sources, mitigation, AI policies, Moral Machine
AI ethics is the set of values, principles and practices that guide the design, development and use of AI so that it benefits people and does not harm them. As AI decides loans, jobs, medical care and news feeds, ethics is essential.
humans can understand why the AI gave a result
no unfair discrimination against any group
works reliably and safely, resists attacks
open about data, purpose and limitations
personal data is protected and used with consent
| Pillar | Question to ask | Example of failure |
|---|---|---|
| Explainability | Why was my loan rejected? | a "black box" gives no reason |
| Fairness | Is any group treated worse? | hiring tool prefers men |
| Robustness | Can it be fooled or fail? | stickers on a stop sign confuse a car |
| Transparency | Do users know AI is used? | hidden AI calls pretending to be human |
| Privacy | Is data collected with consent? | face data scraped without permission |
AI bias is a systematic error that leads to unfair outcomes for certain people or groups.
| Source | Description | Example |
|---|---|---|
| Data bias | training data does not represent everyone | face dataset mostly of light-skinned faces |
| Historical bias | data reflects past discrimination | old hiring data favours men |
| Algorithmic bias | design choices or objectives create unfairness | optimising only for clicks spreads extreme content |
| Human / labelling bias | people labelling data bring prejudice | labellers mark certain accents as "unclear" |
| Sampling bias | data collected only from some places | health app data only from cities |
A company trained an AI on ten years of resumes, mostly from men. The model learned to lower the score of resumes with words like "women's chess club". The company stopped using it. Lesson: check data and test outcomes for every group.
| Issue | Concern |
|---|---|
| Deepfakes and misinformation | fake videos and news |
| Job displacement | need for reskilling |
| Surveillance | loss of privacy and freedom |
| Accountability | who is responsible when AI causes harm? |
| Autonomous weapons | decisions about life and death |
| Environmental cost | large models use much energy and water |
| Policy / framework | Key idea |
|---|---|
| NITI Aayog - National Strategy for AI (#AIforAll), 2018; Responsible AI principles, 2021 | safety, equality, inclusivity, privacy, transparency, accountability |
| Digital Personal Data Protection Act, 2023 (India) | consent and protection of personal data |
| OECD AI Principles, 2019 | human-centred, transparent, robust, accountable AI |
| UNESCO Recommendation on the Ethics of AI, 2021 | global ethical standard |
| EU AI Act | risk-based rules: unacceptable, high, limited, minimal risk |
Using AI to understand a topic and then writing in your own words is responsible. Copying AI output as your own assignment, sharing friends' photos with apps without consent, or making fake images of people is unethical.
Explainability, fairness, robustness, transparency and privacy.
AI bias is a systematic error producing unfair results for some groups. Data bias - unrepresentative data (faces mostly of one skin tone); historical bias - past discrimination in data (old hiring records); algorithmic bias - design choices causing unfairness (optimising only clicks); human labelling bias - prejudice of labellers.
Ethics is a set of moral principles about right and wrong. AI ethics makes sure AI is developed and used in ways that are fair, safe and respect human rights.
| Pillar | Meaning | Example |
|---|---|---|
| Explainability | AI can explain why it made a decision | loan app shows reasons for rejection |
| Fairness | equal treatment of all individuals and groups | hiring AI rates men and women equally |
| Robustness | reliable and secure even with unexpected or attacking inputs | self-driving car safe in fog |
| Transparency | openness about data, design and limits | company publishes how its AI works |
| Privacy | protects personal data and user control | consent before using health data |
| Source of bias | Example |
|---|---|
| Data (sampling) bias | face recognition trained mostly on light skin performs poorly on dark skin |
| Historical bias | hiring AI trained on past hiring (mostly men) rejects women |
| Human / labelling bias | annotators' stereotypes enter the labels |
| Algorithmic bias | design gives too much weight to a proxy like pin code |
| Feedback loop | predictive policing sends more police to the same areas, creating more records there |
Knowing that bias can exist and checking for it at every stage of the AI project cycle.
| Policy / law | Key idea |
|---|---|
| EU AI Act | risk-based law: unacceptable (banned), high, limited, minimal risk |
| NITI Aayog: Responsible AI for All | India's principles: safety, equality, inclusivity, privacy, transparency, accountability |
| DPDP Act 2023 | India's Digital Personal Data Protection Act: consent and protection of personal data |
| UNESCO Recommendation (2021) | global ethical standard on AI adopted by member states |
| OECD AI Principles (2019) | trustworthy AI that respects human rights and democracy |
MIT game: whom should a self-driving car with failed brakes save? It shows moral dilemmas and that people's choices differ.
You hire candidates quickly, then an AI is trained on your choices and copies your bias - showing how biased data creates unfair AI.
Practical: summarise the video "Humans Need Not Apply" (automation and jobs); role-play on biased AI systems; compare AI policies of organisations and governments.
1026 questions in 18 types, including 301 numericals and Python questions with full working
Tap a type to open it. In the one-mark types, tap an option to check yourself. Every answer has an explanation; numericals and Python outputs were worked out by a program. Tap Run on any program to try it in the built-in Python.
Which of these is NOT a method of communication?
(b) Encryption The three methods are verbal, non-verbal and visual.
A presentation that uses charts and infographics is mainly
(b) visual communication Visual communication uses images, charts, signs and symbols.
Body posture, gestures and facial expressions form
(a) non-verbal communication Messages without words are non-verbal.
A person who speaks openly and honestly while respecting others has the
(d) assertive style Assertive is the healthiest style.
A person who agrees on the face but later complains behind the back shows the
(b) passive-aggressive style Anger is expressed indirectly.
A person who always gives in to others, avoids conflict and puts others' needs first shows the
(c) submissive style Submissive communicators fail to express their own needs.
Speaking loudly, interrupting and using blaming words shows the
(a) aggressive style Aggressive communication ignores others' rights.
"I feel upset when the work is late; can we plan better?" This 'I' statement is typical of
(a) assertive communication 'I' statements express feelings without blaming.
A group of words that makes complete sense and has a subject and a verb is a
(a) sentence A phrase does not make complete sense.
'In the morning' is an example of a
(d) phrase It has no subject-verb pair.
Which kind of sentence is "Could you please send the report?"
(c) Interrogative It asks a question and ends with '?'.
In "The team won the match", the subject is
(c) The team The subject is the doer of the action.
Which part of speech joins words or sentences, like 'and', 'but', 'because'?
(b) Conjunction Conjunctions connect.
Which word is an adverb in "She speaks softly"?
(c) softly Adverbs describe verbs.
'Wow!', 'Oh!' and 'Hurray!' are
(d) interjections Interjections express sudden feelings.
Which article is correct: ___ honest man?
(b) an 'Honest' begins with a vowel sound.
Which article is correct: ___ university?
(c) a 'University' begins with a 'yu' (consonant) sound.
'The' is called the
(c) definite article 'A' and 'an' are indefinite articles.
The sentence that states the main idea of a paragraph is the
(d) topic sentence Supporting sentences explain it; the concluding sentence sums up.
A good paragraph has
(d) one main idea with supporting sentences in a logical order Unity and coherence make a good paragraph.
Which preposition fits: The book is ___ the table.
(c) on On shows position on a surface.
Which of these is a visual communication tool in a workplace?
(c) A fire-exit sign Signs and symbols are visual.
Grooming means
(c) taking care of one's appearance: clean, neat and suitably dressed Grooming = presentable appearance and hygiene.
For a job interview, the best dress is
(c) clean, ironed, formal clothes Appropriate dressing creates a good first impression.
A personal grooming checklist helps to
(d) check daily hygiene and appearance habits It lists items like bath, nails, hair, clean clothes, shoes.
Standing straight, smiling and keeping eye contact is
(b) positive body language Positive body language shows confidence.
Folding the arms and avoiding eye contact usually shows
(c) negative or closed body language It may signal defensiveness or nervousness.
Looking within to know our interests, values and abilities is called
(d) self-exploration Techniques: self-reflection, feedback, personality tests, journaling.
Which is a technique of self-exploration?
(c) Keeping a journal and reflecting on experiences Reflection helps us know ourselves.
A group of people working together towards a common goal is a
(b) team Teams share goals and responsibilities.
Which factor helps team building?
(b) Clear goals and mutual trust Trust, clear roles and communication build teams.
Which is NOT a quality of a good team member?
(c) Taking all the credit alone Good members share credit.
The factors that influence teamwork include
(b) communication, trust, leadership and clear roles These decide how well members cooperate.
Time management means
(d) planning and controlling how we use our time to complete tasks It balances work and leisure.
Breaking a large task into smaller tasks helps to
(a) make it manageable and track progress Small steps are easier to schedule.
Putting tasks in order of importance and urgency is called
(b) prioritising Important and urgent tasks come first.
Delaying important tasks again and again is called
(a) procrastination It is the biggest enemy of time management.
Which tool helps in time management?
(a) A to-do list and a calendar app Lists, schedules and planners help.
In the Eisenhower (urgent-important) matrix, tasks that are important and urgent should be
(d) done first Important-urgent tasks are top priority.
Balancing study, play and rest is called
(c) work-life (study-leisure) balance Balance keeps us healthy and productive.
Which SMART letter means the goal must have a deadline?
(d) T - Time-bound Specific, Measurable, Achievable, Realistic, Time-bound.
Which is a sign of good personal grooming?
(b) Trimmed nails and combed hair Neatness shows self-respect.
A team member who listens, resolves conflicts and motivates others shows
(d) leadership Leaders guide and support the team.
Software used to create, edit, format and print text documents is a
(a) word processor Examples: LibreOffice Writer, MS Word, Google Docs.
Which is a word processing software?
(c) LibreOffice Writer Calc is a spreadsheet; Impress makes presentations.
The default file extension of a LibreOffice Writer document is
(b) .odt .odt = OpenDocument Text; MS Word uses .docx.
The shortcut to save a document is
(b) Ctrl + S Ctrl + P prints, Ctrl + O opens, Ctrl + N creates new.
The shortcut to make selected text bold is
(d) Ctrl + B Ctrl + I italic, Ctrl + U underline, Ctrl + E centre.
Which alignment makes text even on both the left and right margins?
(d) Justify Justified text is used in newspapers and books.
Text that appears at the top of every page is the
(a) header The footer appears at the bottom.
Page numbers are usually inserted in the
(a) header or footer They repeat on every page.
The feature that automatically corrects common typing mistakes like 'teh' to 'the' is
(d) AutoCorrect AutoCorrect fixes errors as you type.
Which feature changes every 'colour' to 'color' in a long document in one step?
(b) Find and Replace Shortcut Ctrl + H in many word processors.
How text flows around an inserted picture is controlled by
(d) text wrapping Options: square, tight, in line, behind text.
Lists with 1, 2, 3 or a, b, c are
(b) numbered lists Bulleted lists use dots or symbols.
Arial, Times New Roman and Calibri are names of
(b) fonts (typefaces) Font face, size and style change the look of text.
A grid of rows and columns inserted in a document is a
(b) table Tables organise data neatly.
To save a Writer file so that MS Word users can open it, choose
(d) Save As with the .docx format Documents can be saved in many formats: .odt, .docx, .pdf.
To create a PDF of a document in LibreOffice Writer, use
(a) Export as PDF File > Export as PDF.
Print Preview is used to
(d) see how the document will look when printed It saves paper by checking layout first.
The shortcut to undo the last action is
(c) Ctrl + Z Ctrl + Y redoes; Ctrl + A selects all.
Portrait and landscape are types of
(b) page orientation Landscape is wider than it is tall.
Which key combination selects the entire document?
(c) Ctrl + A A for All.
Values are
(a) beliefs that guide our behaviour and decisions Values tell us what is important and right.
Which is an entrepreneurial value?
(c) Innovativeness Entrepreneurial values: innovativeness, independence, outstanding performance, respect for work.
An entrepreneur who likes to make his or her own decisions and does not depend on others shows
(d) independence Independence is an entrepreneurial value.
Always trying to do the best and set high standards is the value of
(d) outstanding performance Entrepreneurs aim for excellence.
Believing that no work is small and doing every task with dignity is
(a) respect for work Respect for all kinds of work.
Creating new products or new ways of doing things is
(b) innovativeness Innovation drives business growth.
An attitude is
(c) a way of thinking or feeling about something that shows in behaviour Attitudes can be positive or negative.
Entrepreneurs prefer to take
(d) moderate (calculated) risks They study the situation and then take sensible risks.
'Believing that you can change the environment' is
(c) an entrepreneurial attitude Entrepreneurs believe they can shape their surroundings.
Seeing a business chance where others see only a problem is
(a) seeking economic opportunities Opportunity-seeking is a key attitude.
Using imagination and gut feeling to find new ideas is
(d) imagination and intuition Entrepreneurs use both imagination and analysis.
Studying a situation and making a plan before acting is the attitude of
(c) analysing situations and planning action Planning reduces risk.
Which is NOT an entrepreneurial attitude?
(d) Waiting for others to tell you what to do Entrepreneurs take initiative.
An entrepreneur who enjoys the freedom to express ideas and act on them shows
(c) enjoying freedom of expression and action Freedom is a strong motivator.
Which of these shows the attitude 'getting involved in activities'?
(c) Volunteering to organise the school fair stall Entrepreneurs participate actively.
A student who repairs old phones and sells them, instead of waiting for a job, shows
(b) an entrepreneurial attitude Taking initiative to create value.
Attitudes can be changed by
(b) learning, experience and practice Entrepreneurial attitudes can be developed.
Which value is shown by an entrepreneur who keeps improving the product after customer feedback?
(d) Outstanding performance and innovativeness Continuous improvement shows these values.
A positive attitude in business helps to
(c) face failures and keep trying Positive people bounce back.
Which is an example of a moderate risk?
(a) Starting a small tiffin service after a market survey Moderate risks are studied and planned.
A green economy is
(a) low-carbon, resource-efficient and socially inclusive UNEP definition.
Which is a main sector of the green economy?
(c) Renewable energy Main sectors: e-waste management, green transport, renewable energy, green construction, water management.
Safe collection and recycling of old phones and computers is
(a) e-waste management E-waste has toxic metals like lead and mercury.
Electric buses, metro rail and cycling are examples of
(d) green transportation They reduce fuel use and pollution.
Buildings that use less energy and water and use eco-friendly materials are
(a) green buildings (green construction) They follow rating systems like GRIHA and IGBC.
Solar, wind and small hydro power are
(d) renewable energy sources They are replenished naturally.
Drip irrigation and rainwater harvesting belong to
(d) water management They save water.
India's mission to keep cities and villages clean is
(b) Swachh Bharat Mission A key policy initiative for greening.
The National Action Plan on Climate Change (NAPCC) has missions such as
(b) the National Solar Mission NAPCC (2008) has eight national missions including solar and energy efficiency.
Which scheme promotes electric vehicles in India?
(d) FAME (Faster Adoption and Manufacturing of Electric Vehicles) FAME supports EVs.
People or groups who have an interest in the green economy, such as government, industry and citizens, are
(a) stakeholders Stakeholders: government, private sector, NGOs, communities, consumers.
Who makes laws and policies like pollution standards for the green economy?
(a) Government Government regulates and gives incentives.
Private companies help greening by
(d) investing in clean technology and reducing their emissions Corporate social responsibility and green investment.
Tourism that conserves nature and benefits local people is
(b) eco-tourism Green tourism protects the environment.
Organic farming and reduced use of chemical fertilisers come under greening of
(b) agriculture Sustainable agriculture protects soil.
Avoiding overfishing and protecting breeding seasons is part of greening
(b) fisheries Sustainable fisheries keep fish stocks alive.
LED bulbs and 5-star rated appliances help in
(d) energy efficiency They use less electricity.
Planting trees on barren land is called
(a) afforestation It increases forest cover.
Which is a role of citizens in the green economy?
(c) Reducing, reusing and recycling waste Every person is a stakeholder.
Green jobs are jobs that
(d) help protect or restore the environment E.g. solar technician, recycling worker, organic farmer.
In the CBSE practical activity, applications are categorised into
(b) the three domains of AI Data Science, CV and NLP.
A neural network is inspired by
(c) the human brain Neurons connected in layers.
AI that would be smarter than humans in every field is
(d) Super AI It is only a theoretical idea.
Data used to teach a model is
(b) training data Testing data checks the trained model.
An AI that has been trained only on photos of cats and dogs is shown a horse. It will most likely
(a) wrongly call it a cat or a dog AI only knows what it was trained on.
IBM Watson became famous in 2011 by
(d) winning the quiz show Jeopardy! It understood natural-language questions.
Machine learning that uses neural networks with many layers is
(c) Deep Learning Deep = many hidden layers.
IBM's Deep Blue became famous in 1997 for
(b) defeating world chess champion Garry Kasparov A landmark of AI in games.
In AI terminology, the input variables used to make a prediction are called
(d) features The value to be predicted is the label (target).
Finding hidden groups in unlabelled data is
(b) unsupervised learning Example: customer segmentation.
AI that has its own consciousness and self-awareness is
(c) self-aware AI It is only hypothetical.
The test proposed in 1950 to check whether a machine can show human-like intelligence is the
(a) Turing Test Alan Turing, 'Computing Machinery and Intelligence' (1950).
AlphaGo, which defeated Go champion Lee Sedol in 2016, was made by
(d) Google DeepMind It used deep reinforcement learning.
Predicting house prices from past sales data belongs to
(d) Data Science It works on numbers in tables.
The value that a supervised model learns to predict is called the
(a) label (target) Features -> model -> label.
Self-driving cars that use recent observations of other cars to decide are
(d) limited memory AI They store data for a short time.
Learning from labelled data (input with the correct output) is
(b) supervised learning Example: emails labelled spam / not spam.
A system that learns from data without being explicitly programmed is
(b) Machine Learning Arthur Samuel's definition.
AI that could do any intellectual task a human can do is
(d) General (Strong) AI It does not exist yet.
Reading number plates of vehicles at a toll plaza uses
(c) Computer Vision It processes images and video.
Sentiment analysis of product reviews belongs to
(c) Natural Language Processing It works on text.
Who is called the 'father of Artificial Intelligence' and coined the term in 1956?
(d) John McCarthy At the Dartmouth Conference, 1956.
The 1956 workshop considered the birth of AI as a field was held at
(b) Dartmouth College Dartmouth Summer Research Project on AI.
An agent that learns by getting rewards and penalties is using
(c) reinforcement learning Example: game-playing AI, robots.
Based on functionality, AI that only reacts to the present situation and has no memory (like Deep Blue) is a
(a) reactive machine It cannot use past experiences.
The three domains of AI are
(b) Data Science, Computer Vision and Natural Language Processing Based on the type of data used.
The three types of layers in a neural network are
(a) input, hidden and output Hidden layers do the processing.
Artificial Intelligence is best defined as
(d) the ability of machines to perform tasks that normally need human intelligence Learning, reasoning, perception and decision making.
Which is a benefit of AI?
(c) It can work 24x7 on repetitive tasks with high accuracy AI reduces human error in repetitive work.
AI that could understand human emotions, beliefs and intentions is called
(d) theory of mind AI It is still being researched.
Which is a limitation of AI?
(a) It lacks common sense and true understanding AI cannot think like humans in new situations.
ChatGPT, released in 2022, is an example of
(a) a Large Language Model (generative AI) It generates human-like text.
Recommendation of videos on YouTube is an application of
(a) Machine Learning on user data It learns your viewing patterns.
Periods when funding and interest in AI fell sharply are called
(c) AI winters There were AI winters in the 1970s and late 1980s.
ELIZA (1966) was one of the first
(d) chatbots Created by Joseph Weizenbaum at MIT.
AI that is designed for one specific task, like face unlock, is
(d) Narrow (Weak) AI All AI in use today is narrow AI.
Computing that tries to mimic human thinking - perception, learning and reasoning - is called
(b) cognitive computing IBM Watson is an example.
IBM SkillsBuild is
(a) a free online learning platform with AI courses and badges CBSE uses it for certification in AI 843.
In the practical activity, students identify
(b) ten companies hiring for specific AI positions From job portals like LinkedIn and Naukri.
An engineer who builds face recognition and medical image systems specialises in
(d) Computer Vision CV engineers work with images and video.
Labelling images so that an AI model can learn from them is done by
(d) data annotators (labellers) An entry-level AI job.
Which is a soft skill needed in AI careers?
(c) Communication and teamwork AI professionals must explain results and work in teams.
Who develops new AI algorithms and publishes research papers?
(b) AI Research Scientist Research scientists push the field forward.
AI in healthcare is used for
(b) reading X-rays and predicting diseases Medical imaging and diagnosis.
AI in banking and finance is used for
(b) fraud detection and credit scoring It spots unusual transactions.
AI in transportation is used for
(c) self-driving cars and traffic prediction Route optimisation and autonomous vehicles.
A person who makes sure AI systems are fair, transparent and responsible is an
(b) AI Ethicist AI ethics is a growing career.
A person who studies data and prepares reports and dashboards for business decisions is a
(b) Data Analyst Analysts answer business questions with data.
AI in manufacturing is used for
(a) predictive maintenance and quality inspection It predicts machine failure and finds defects.
TensorFlow and PyTorch are
(c) deep learning frameworks They build and train neural networks.
Which programming language is most widely used in AI and data science?
(c) Python Python has rich AI libraries.
Which is a technical skill for an AI career?
(c) Programming and data handling Also ML, statistics and tools.
Which mathematical areas are essential for AI careers?
(b) Statistics, probability, linear algebra and calculus They are the base of ML algorithms.
A good first step to choose an AI career path is to
(b) match your interests and skills with job roles Self-assessment is part of the unit.
Who designs, builds and deploys machine learning models in products?
(b) Machine Learning Engineer ML engineers focus on building and scaling models.
An engineer who builds chatbots, translators and speech systems specialises in
(d) Natural Language Processing NLP engineers work with text and speech.
Jupyter Notebook is used for
(b) writing and running code with notes and charts together Popular among data scientists.
Demand for AI professionals worldwide is
(b) growing rapidly across many industries AI is used in health, finance, retail, farming and more.
MLOps engineers mainly work on
(a) deploying, monitoring and maintaining ML models in production Ops = operations.
Which job role needs strong statistics and visualisation skills to turn data into stories for managers?
(a) Data Analyst Analysts communicate insights.
Which statement about AI careers is TRUE?
(b) People from many backgrounds (biology, finance, arts) can work in AI with the right skills AI is interdisciplinary.
A professional who decides what an AI product should do and coordinates the team is an
(a) AI Product Manager Product managers link business needs and technology.
Which online resource gives free datasets and competitions for AI learners?
(a) Kaggle Kaggle hosts datasets and ML contests.
A professional who collects, cleans and analyses data to find insights and build models is a
(d) Data Scientist Data scientists combine statistics, programming and domain knowledge.
AI in retail and e-commerce is used for
(b) product recommendations and demand forecasting Personalised shopping.
Which skill helps an AI professional keep up with fast changes in the field?
(c) Continuous (lifelong) learning AI tools change very fast.
Scikit-learn is a Python library mainly used for
(d) machine learning (classification, regression, clustering) It has ready ML algorithms.
AI in agriculture is used for
(b) crop disease detection and yield prediction Drones, sensors and image analysis help farmers.
A professional who designs robots that sense and act in the real world is a
(c) Robotics Engineer Robotics combines AI, mechanics and electronics.
What is the output of the following code?
s = "deep learning"
print(s.upper(), s.title(), s.count("a"))(c) DEEP LEARNING Deep Learning 1 upper(), title() and count() are string methods.
What is the output of the following code?
s = "data science"
print(s.upper(), s.title(), s.count("a"))(d) DATA SCIENCE Data Science 2 upper(), title() and count() are string methods.
What is the output of the following code?
for i in range(5, 8, 2):
print(i, end=' ')(d) 5 7 range(5, 8, 2) starts at 5, adds 2 each time and stops before 8.
What is the output of the following code?
t = (10, 20, 30, 20) print(t.count(20), t.index(30), t[-1])
(d) 2 2 20 count gives occurrences; index gives the first position.
Which is a valid identifier?
(a) _rate2 Cannot start with a digit, contain '-', or be a keyword.
What is the output of the following code?
s = 0
for i in range(1, 4):
s = s + i
print(s)(d) 6 It adds 1 to 3: sum = 6.
The data type of 3 + 4j is
(b) complex j denotes the imaginary part.
What is the output of the following code?
a = "KV" b = "3" print(a + b)
(c) KV3 Both are strings, so + joins them.
What is the output of the following code?
i = 1
while i < 40:
print(i, end=' ')
i = i * 2(b) 1 2 4 8 16 32 i is multiplied by 2 each round while it is less than 40.
What is the output of the following code?
x = 12 x += 4 x *= 4 print(x)
(b) 64 x = 12, then x = 12 + 4 = 16, then x = 16 * 4 = 64.
Which library provides ready machine learning algorithms like LinearRegression and KNeighborsClassifier?
(b) scikit-learn (sklearn) Scikit-learn is the main ML library in Python.
The else block of a for loop runs when
(a) the loop ends normally without break A loop-else runs if no break occurred.
What is the output of the following code?
marks = 91
if marks >= 75:
print('Distinction')
elif marks >= 33:
print('Pass')
else:
print('Fail')(b) Distinction 91 is at least 75, so that branch runs and the rest are skipped.
What is the output of the following code?
for i in range(5, 13, 3):
print(i, end=' ')(c) 5 8 11 range(5, 13, 3) starts at 5, adds 3 each time and stops before 13.
What is the output of the following code?
print(type("25"))(b) <class 'str'> The value "25" is of type str.
What is the output of the following code?
a = 22 b = 4 print(a > b and b > 5, a > b or b > 5, not a > b)
(d) False True False a > b is True, b > 5 is False; and needs both True, or needs one.
Which function creates a NumPy array from a list?
(a) np.array() import numpy as np.
What is the output of the following code?
L = [50, 35, 95, 30, 40] print(L[-1], L[-2], len(L))
(d) 40 30 5 -1 is the last, -2 the second last; len gives 5.
What is the output of the following code?
n = "9" print(n * 2)
(b) 99 n is a string, so * 2 repeats it.
Which of these is NOT a token type in Python?
(b) Comment Comments are ignored by the interpreter.
What is the output of the following code?
L = [1, 2, 3] M = L M.append(4) print(L)
(b) [1, 2, 3, 4] M and L refer to the same list.
What is the output of the following code?
a = 2 b = 10 print(a > b and b > 5, a > b or b > 5, not a > b)
(c) False True True a > b is False, b > 5 is True; and needs both True, or needs one.
What is the output of the following code?
L = [85, 50, 55, 30, 15] L.insert(1, 0) L.remove(15) print(L)
(b) [85, 0, 50, 55, 30] insert(1, 0) puts 0 at index 1; remove(15) deletes that value.
Symbols like ( ) [ ] { } , : are
(c) punctuators (delimiters) They organise code structure.
Pandas is mainly used for
(d) data analysis with Series and DataFrames Tabular data handling.
What is the output of the following code?
marks = 53
if marks >= 75:
print('Distinction')
elif marks >= 33:
print('Pass')
else:
print('Fail')(b) Pass 53 is at least 33 but less than 75, so that branch runs and the rest are skipped.
Python is case-sensitive. This means
(c) Marks and marks are different names Case matters in identifiers.
What is the output of the following code?
print(6 // 4, 6 % 4)
(c) 1 2 6 // 4 is the floor quotient 1; 6 % 4 is the remainder 2.
What is the output of the following code?
import numpy as np a = np.array([10, 20, 30, 40]) print(a[a > 15].mean())
(c) 30.0 Values above 15: 20, 30, 40; mean = 30.0.
What is the output of the following code?
i = 1
while i < 40:
print(i, end=' ')
i = i * 3(d) 1 3 9 27 i is multiplied by 3 each round while it is less than 40.
What is the output of the following code?
L = [25, 20, 75, 90, 60] L.append(7) print(L)
(b) [25, 20, 75, 90, 60, 7] append() adds 7 at the end.
Reserved words with a special meaning in Python, like if, for, while, are
(d) keywords They cannot be used as names.
Names given to variables, functions and classes are
(c) identifiers e.g. total_marks.
What is the output of the following code?
x = 14 x += 6 x *= 3 print(x)
(b) 60 x = 14, then x = 14 + 6 = 20, then x = 20 * 3 = 60.
What is the output of the following code?
x = 5 x = x * 2 ** 2 - 3 print(x)
(b) 17 first: 2 2 = 4; then 5 * 4 = 20; 20 - 3 = 17.
What is the output of the following code?
print("AI" * 3)(a) AIAIAI * repeats the string 3 times.
Which has the highest precedence?
(b) ** Exponent comes before multiplication and addition.
What is the output of the following code?
s = "Artificial" print(s[2:5], s[-3:], s[::2])
(c) tif ial Atfca Slicing: start included, stop excluded; ::2 takes every second character.
What is the output of the following code?
s = {3, 1, 3, 2, 1}
print(len(s), sorted(s))(a) 3 [1, 2, 3] Sets keep only unique values.
What is the output of the following code?
s = "Computer" print(s[2:5], s[-3:], s[::2])
(c) mpu ter Cmue Slicing: start included, stop excluded; ::2 takes every second character.
The two modes of working in Python IDLE are
(b) interactive mode and script mode Shell (>>>) vs .py file.
What is the output of the following code?
n = 5
c = 0
while n > 0:
n = n - 2
c = c + 1
print(c, n)(b) 3 -1 The loop subtracts 2 until n is 0 or less; count the rounds.
What is the output of the following code?
d = {"a": 1, "b": 2}
d["c"] = 3
print(len(d), d.get("b"), d.get("z", 0))(b) 3 2 0 Adding a key; get() returns the default if the key is missing.
What is the output of the following code?
import numpy as np a = np.arange(1, 7).reshape(2, 3) print(a.shape, a[1][2])
(d) (2, 3) 6 arange(1, 7) gives 1 to 6; reshaped into 2 rows and 3 columns.
What is the output of the following code?
print(type([2, 5]))
(d) <class 'list'> The value [2, 5] is of type list.
What is the output of the following code?
L = [5, 3, 8] print(L * 2, 8 in L)
(c) [5, 3, 8, 5, 3, 8] True * repeats a list; in checks membership.
What is the output of the following code?
import pandas as pd
df = pd.DataFrame({"x": [1, 2, 3], "y": [4, 5, 6]})
print(df.shape, df["y"].sum())(d) (3, 2) 15 3 rows and 2 columns; 4 + 5 + 6.
What is the output of the following code?
print(type(25.0))
(d) <class 'float'> The value 25.0 is of type float.
Which statement is used to exit a loop early?
(a) break continue skips to the next iteration.
Which statement does nothing and is used as a placeholder?
(b) pass Used for empty blocks.
What is the output of the following code?
n = 10
c = 0
while n > 0:
n = n - 3
c = c + 1
print(c, n)(c) 4 -2 The loop subtracts 3 until n is 0 or less; count the rounds.
What is the output of the following code?
marks = 90
if marks >= 75:
print('Distinction')
elif marks >= 33:
print('Pass')
else:
print('Fail')(d) Distinction 90 is at least 75, so that branch runs and the rest are skipped.
The value None in Python represents
(c) the absence of a value Its type is NoneType.
What is the output of the following code?
n = "38" print(int(n) + 2)
(d) 40 int("38") converts the string to the number 38; 38 + 2 = 40.
What is the output of the following code?
L = [x * x for x in range(1, 5) if x % 2 == 0] print(L)
(a) [4, 16] Even x: 2 and 4; squares 4 and 16.
What is the output of the following code?
import pandas as pd s = pd.Series([4, 9, 2], index=["a", "b", "c"]) print(s["b"], s.max(), s.idxmin())
(a) 9 9 c Label b holds 9; idxmin gives the label of the smallest value.
What is the output of the following code?
marks = 82
if marks >= 75:
print('Distinction')
elif marks >= 33:
print('Pass')
else:
print('Fail')(d) Distinction 82 is at least 75, so that branch runs and the rest are skipped.
What is the output of the following code?
print(type(25))
(c) <class 'int'> The value 25 is of type int.
np.zeros((2, 3)) creates
(d) a 2 x 3 array filled with 0.0 zeros(shape).
What is the output of the following code?
import numpy as np a = np.array([2, 4, 6]) print(a * 2, a.sum())
(c) [ 4 8 12] 12 NumPy multiplies element-wise.
What is the output of the following code?
a = "AI" b = "4" print(a + b)
(b) AI4 Both are strings, so + joins them.
What is the output of the following code?
a = 29 b = 21 print(a > b and b > 5, a > b or b > 5, not a > b)
(d) True True False a > b is True, b > 5 is True; and needs both True, or needs one.
What is the output of the following code?
for i in range(5):
if i == 3:
break
print(i, end=" ")(c) 0 1 2 break stops the loop at i = 3.
Which Python module is used to read and write CSV files?
(a) csv csv.reader and csv.writer.
What is the output of the following code?
print("AI" * 2)(d) AIAI * repeats the string 2 times.
What is the output of the following code?
L = [80, 95, 50, 75, 70] print(L[1:4])
(c) [95, 50, 75] Slice 1:4 gives indexes 1, 2 and 3.
What is the output of the following code?
marks = 91
if marks >= 75:
print('Distinction')
elif marks >= 33:
print('Pass')
else:
print('Fail')(d) Distinction 91 is at least 75, so that branch runs and the rest are skipped.
Which Pandas function reads a CSV file into a DataFrame?
(c) pd.read_csv() e.g. pd.read_csv('rainfall.csv').
What is the output of the following code?
L = [55, 40, 90, 70, 45] print(L[1:4])
(c) [40, 90, 70] Slice 1:4 gives indexes 1, 2 and 3.
A two-dimensional labelled table in Pandas is a
(b) DataFrame A Series is one-dimensional.
What is the output of the following code?
s = "Analytics" print(s[2:5], s[-3:], s[::2])
(c) aly ics Aayis Slicing: start included, stop excluded; ::2 takes every second character.
What is the output of the following code?
s = "Vision" print(len(s), s[0], s[-1])
(d) 6 V n len counts 6 characters; index 0 is the first and -1 the last character.
Which data type is immutable?
(d) tuple Tuples cannot be changed after creation.
Which data type stores only unique values with no order?
(a) set Sets remove duplicates.
What is the output of the following code?
L = [40, 95, 75, 85, 20] L.sort() print(L)
(d) [20, 40, 75, 85, 95] sort() arranges in ascending order.
train_test_split() is used to
(c) divide data into training and testing sets From sklearn.model_selection.
What is the output of the following code?
for i in range(4, 9, 2):
print(i, end=' ')(c) 4 6 8 range(4, 9, 2) starts at 4, adds 2 each time and stops before 9.
What is the output of the following code?
n = "13" print(int(n) + 2)
(c) 15 int("13") converts the string to the number 13; 13 + 2 = 15.
What is the output of the following code?
L = [50, 70, 25, 40, 35] L.sort() print(L)
(c) [25, 35, 40, 50, 70] sort() arranges in ascending order.
What is the output of the following code?
print(16 + 2 * 4)
(b) 24 * is done before +: 2 * 4 = 8; 16 + 8 = 24.
In scikit-learn, the method used to get predictions from a trained model is
(a) predict() model.predict(X_test).
{'a': 1, 'b': 2} is a
(c) dictionary Key-value pairs in braces.
What is the output of the following code?
print(9 + 2 * 4)
(d) 17 * is done before +: 2 * 4 = 8; 9 + 8 = 17.
Python was developed by
(a) Guido van Rossum First released in 1991.
What is the output of the following code?
i = 1
while i < 20:
print(i, end=' ')
i = i * 2(d) 1 2 4 8 16 i is multiplied by 2 each round while it is less than 20.
What is the output of the following code?
print(17 % 5 * 2 + 10 // 4)
(d) 6 17 % 5 = 2; 2 * 2 = 4; 10 // 4 = 2; 4 + 2 = 6.
In scikit-learn, the method used to train a model is
(d) fit() model.fit(X_train, y_train).
What is the output of the following code?
print(not 5 > 3 or 2 > 1 and 4 < 2)
(c) False not first, then and, then or: False or (True and False) = False.
What is the output of the following code?
L = [80, 85, 20, 50, 35] print(L[-1], L[-2], len(L))
(b) 35 50 5 -1 is the last, -2 the second last; len gives 5.
What is the output of the following code?
s = 0
for i in range(1, 4):
s = s + i
print(s)(a) 6 It adds 1 to 3: sum = 6.
Fixed values such as 25, 3.14 and 'AI' are
(c) literals Numeric, string and boolean literals.
What is the output of the following code?
for i in range(5):
if i == 3:
continue
print(i, end=" ")(b) 0 1 2 4 continue skips only i = 3.
What is the output of the following code?
a = 11 b = 24 print(a > b and b > 5, a > b or b > 5, not a > b)
(c) False True True a > b is False, b > 5 is True; and needs both True, or needs one.
df.head() shows
(c) the first 5 rows of the DataFrame head(n) shows the first n rows.
What is the output of the following code?
n = "3" print(n * 2)
(d) 33 n is a string, so * 2 repeats it.
Which operator checks whether a value is present in a list?
(b) in Membership operators: in, not in.
What is the output of the following code?
print(3 ** 2, 65 / 5)
(b) 9 13.0 3 ** 2 = 9; / always gives a float: 65 / 5 = 13.0.
What is the output of the following code?
L = [55, 70, 45, 65, 15] L.append(7) print(L)
(c) [55, 70, 45, 65, 15, 7] append() adds 7 at the end.
What is the output of the following code?
s = "python ai"
print(s.upper(), s.title(), s.count("a"))(d) PYTHON AI Python Ai 1 upper(), title() and count() are string methods.
What is the output of the following code?
print(4 ** 2, 40 / 5)
(a) 16 8.0 4 ** 2 = 16; / always gives a float: 40 / 5 = 8.0.
Python's character set includes
(c) letters, digits, special symbols, whitespace and Unicode characters Python 3 supports Unicode.
What is the output of the following code?
s = 0
for i in range(1, 4):
s = s + i
print(s)(a) 6 It adds 1 to 3: sum = 6.
What is the output of the following code?
print(type(3 / 3))
(b) <class 'float'> The value 3 / 3 is of type float.
Python is
(a) an interpreted, high-level, general-purpose language Code runs line by line through the interpreter.
The smallest individual unit of a Python program is a
(c) token Tokens: keywords, identifiers, literals, operators, punctuators.
A CSV file stores data as
(c) plain text with values separated by commas Comma Separated Values; opens in spreadsheets.
Which operator is used for exponent?
(b) ** 2 ** 3 = 8.
The 'is' operator checks whether two variables
(b) refer to the same object Identity operators: is, is not.
What is the output of the following code?
L = [55, 45, 25, 60, 50] L.insert(1, 0) L.remove(50) print(L)
(a) [55, 0, 45, 25, 60] insert(1, 0) puts 0 at index 1; remove(50) deletes that value.
What is the output of the following code?
s = "Learning" print(s[2:5], s[-3:], s[::2])
(d) arn ing Lann Slicing: start included, stop excluded; ::2 takes every second character.
NumPy is mainly used for
(b) fast numerical arrays and mathematical operations Numerical Python.
What is the output of the following code?
a = 9 b = 19 print(a > b and b > 5, a > b or b > 5, not a > b)
(c) False True True a > b is False, b > 5 is True; and needs both True, or needs one.
What is the output of the following code?
print(type(True))
(b) <class 'bool'> The value True is of type bool.
What is the output of the following code?
print(8 // 6, 8 % 6)
(a) 1 2 8 // 6 is the floor quotient 1; 8 % 6 is the remainder 2.
What is the output of the following code?
s = "Data" print(len(s), s[0], s[-1])
(b) 4 D a len counts 4 characters; index 0 is the first and -1 the last character.
Which function gives the shape (rows, columns) of a NumPy array a?
(d) a.shape shape is an attribute.
What is the output of the following code?
for i in range(3):
pass
else:
print("done", i)(c) done 2 The loop ended normally, so else runs; i is 2.
In the CBSE capstone project, the first stages are
(d) prepare for the project, form a team and define the problem Then understand users, brainstorm, design the solution.
A short summary of a project giving the problem, users, solution and impact is the
(c) project abstract Written using the design thinking framework.
IBM SkillsBuild's course recommended for this unit is
(d) What is Design Thinking? Listed as the resource in the curriculum.
How many Sustainable Development Goals (SDGs) did the UN adopt in 2015?
(b) 17 To be achieved by 2030.
A capstone project is
(b) a final project that applies all the learning to solve a real-world problem In AI 843 it is done in teams and linked to SDGs.
Trying the prototype with real users and collecting feedback is the
(a) Test stage Feedback may send us back to earlier stages.
Writing a clear problem statement based on user needs is the
(c) Define stage Also called the point-of-view statement.
A user persona is
(c) a fictional profile representing a typical user It helps designers think of users.
Brainstorming many possible solutions without judging them is the
(c) Ideate stage Quantity of ideas matters here.
In a capstone project, aligning the problem to an SDG ensures that
(c) the project addresses a real social or environmental need SDG alignment gives purpose.
Which is an example of empathising?
(a) Interviewing elderly people about difficulties in using smartphones It captures user experience.
An AI app that detects crop diseases to help farmers mainly supports
(c) SDG 2: Zero Hunger It improves food production.
The five stages of design thinking in order are
(d) Empathize, Define, Ideate, Prototype, Test Stanford d.school model.
Breaking a big problem into smaller, manageable parts is called
(c) problem decomposition A key step of computational thinking.
Design thinking is iterative. This means
(b) stages can be repeated until the solution works well Learn, improve and repeat.
The 5W1H method asks
(a) Who, What, When, Where, Why and How It breaks a problem into parts.
Why is user feedback important in design thinking?
(c) It shows whether the solution really solves the user's problem Testing with users drives improvement.
The 'How Might We' question is used in
(d) the Define and Ideate stages to frame problems as opportunities e.g. How might we help students manage stress?
Asking 'why' repeatedly to find the root cause of a problem is the
(b) 5 Whys technique It digs to the root cause.
SDG 4 is
(b) Quality Education SDG 2 Zero Hunger, 6 Clean Water, 13 Climate Action.
Building a simple, low-cost model of the solution is the
(d) Prototype stage Paper models, mock-ups, simple apps.
Expressing an AI solution in non-technical terms helps to
(d) explain it to users and stakeholders Good communication is part of the unit.
A prototype should be
(b) quick, cheap and good enough to test the idea Fail fast, learn fast.
An AI tool that predicts air pollution levels mainly supports
(a) SDG 13: Climate Action / SDG 11: Sustainable Cities It helps cities act on pollution.
Which stage of design thinking answers 'Who is the user and what do they need?'
(b) Empathize and Define Understanding before solving.
In an empathy map, "stays up late revising" goes in
(b) Does Actions and behaviour go in Does.
Which tool is commonly used to organise brainstorming ideas?
(d) Mind map or sticky notes Visual tools help ideation.
An empathy map has four quadrants:
(d) Says, Thinks, Does, Feels It captures what users say, think, do and feel.
In an empathy map, "I am worried I will fail the exam" goes in
(b) Feels (or Thinks) It is an inner emotion.
Understanding users' needs and feelings by observing and interviewing them is the
(a) Empathize stage The first stage.
SDG 3 is
(c) Good Health and Well-being SDG 1 No Poverty, 5 Gender Equality, 15 Life on Land.
Design thinking is
(b) a human-centred approach to solving problems creatively It starts by understanding users.
In the CBSE practical, Python programs for visualisation use the dataset
(b) rainfall.csv Listed in the curriculum.
A grayscale image is 28 x 28 pixels. How many values does its pixel matrix hold?
(b) 784 28 x 28 = 784.
The level of measurement of 'Number of children in a family' is
(a) Ratio Equal gaps and a true zero, so ratios make sense.
A 3 x 3 identity matrix has how many 1s?
(c) 3 1s only on the main diagonal.
Data literacy is the ability to
(b) read, understand, analyse and communicate with data It is the base of working with AI.
Find the standard deviation (population) of: 6, 14, 8, 8.
(d) 3 Mean = 9, variance = 9, SD = sqrt(9) = 3.
Using min-max normalisation with min = 0 and max = 50, the value 45 becomes
(d) 0.9 (x - min) / (max - min) = (45 - 0) / (50 - 0) = 0.9.
Census reports, published research and online datasets are
(c) secondary data Already collected by others.
Scaling values to a common range like 0 to 1 is called
(a) normalisation Min-max normalisation: (x - min) / (max - min).
Customer satisfaction (poor, fair, good, excellent) is
(d) ordinal data Ordered categories.
A matrix is
(c) a rectangular arrangement of numbers in rows and columns Written in brackets.
Which Python library function gives the standard deviation of a list x?
(a) statistics.pstdev(x) (population) or numpy.std(x) statistics.stdev uses n - 1 (sample).
A = [0 -1; 6 6] and B = [1 3; 9 -1]. Find A + B.
(b) [1 2; 15 5] Add the corresponding elements.
Height, weight and income are on the
(d) ratio scale They have a true zero, so ratios make sense.
A matrix with the same number of rows and columns is a
(d) square matrix e.g. 2 x 2, 3 x 3.
Two matrices can be added only if
(d) they have the same order Add corresponding elements.
The level of measurement of 'Education level: primary, secondary, graduate' is
(c) Ordinal Categories with an order, but gaps are not equal.
The four levels of measurement are
(a) nominal, ordinal, interval and ratio Stevens' levels of measurement.
The transpose of a matrix is obtained by
(a) interchanging its rows and columns A of order m x n becomes n x m.
Find the mean of: 51, 23, 6, 42, 40, 23, 37.
(c) 31.71 Mean = 222 / 7 = 31.71.
The level of measurement of 'Temperature in degree Fahrenheit' is
(b) Interval Equal gaps but no true zero.
Which chart shows the frequency of values in ranges (bins)?
(c) Histogram Bars touch each other for continuous ranges.
The mean of 4, 8, 6, 10, 2 is
(c) 6 Sum 30 / 5 = 6.
Find the transpose of A = [0 0; 8 -3].
(a) [0 8; 0 -3] Rows become columns.
Using min-max normalisation with min = 0 and max = 50, the value 15 becomes
(d) 0.3 (x - min) / (max - min) = (15 - 0) / (50 - 0) = 0.3.
A dataset of 500 rows is split 75 : 25 into training and testing sets. How many rows are in the testing set?
(b) 125 25% of 500.
Which chart is best for showing parts of a whole?
(a) Pie chart Slices add up to 100%.
A dataset of 2500 rows is split 70 : 30 into training and testing sets. How many rows are in the testing set?
(b) 750 30% of 2500.
The level of measurement of 'IQ score' is
(b) Interval Equal gaps but no true zero.
If all values in a data set are equal, the standard deviation is
(c) 0 There is no spread.
The level of measurement of 'Grades A, B, C, D' is
(b) Ordinal Categories with an order, but gaps are not equal.
Which is an online source of secondary data?
(c) data.gov.in Open Government Data Platform India.
Find the standard deviation (population) of: 19, 13, 19, 13.
(d) 3 Mean = 16, variance = 9, SD = sqrt(9) = 3.
Find the median of: 28, 6, 28, 55, 38, 59.
(a) 33 Sorted: 6, 28, 28, 38, 55, 59; median = 33.
Variance measures
(c) how far the values are spread from the mean Average of squared deviations.
Find the (population) variance of: 8, 2, 10, 19, 16.
(d) 36 Mean = 11. Variance = [(8 - 11)^2 + (2 - 11)^2 + (10 - 11)^2 + (19 - 11)^2 + (16 - 11)^2] / 5 = 180 / 5 = 36.
A square matrix with 1s on the main diagonal and 0s elsewhere is the
(b) identity matrix A x I = A.
Find the transpose of A = [-2 -2; 4 8].
(a) [-2 4; -2 8] Rows become columns.
Using min-max normalisation with min = 20 and max = 120, the value 90 becomes
(c) 0.7 (x - min) / (max - min) = (90 - 20) / (120 - 20) = 0.7.
A grayscale image is stored in a computer as a
(c) matrix of pixel values Linear algebra is used in Computer Vision.
Find the mode of: 7, 48, 48, 42, 52, 30, 16.
(a) 48 48 occurs most often (2 times).
The level of measurement of 'Favourite colour' is
(d) Nominal Only names/categories; no order.
Which chart is best to show the trend of rainfall over 12 months?
(a) Line graph Line graphs show change over time.
The level of measurement of 'Calendar year (2010, 2020)' is
(d) Interval Equal gaps but no true zero.
Blood group (A, B, AB, O) is measured on the
(c) nominal scale Names only, no order.
The level of measurement of 'Monthly income in rupees' is
(b) Ratio Equal gaps and a true zero, so ratios make sense.
A is 2 x 3 and B is 2 x 3. Is AB defined?
(a) No, columns of A (3) are not equal to rows of B (2) Multiplication needs columns of A = rows of B.
In matplotlib, the function to draw a bar chart is
(c) plt.bar() plt.plot() draws lines.
A matrix has 2 rows and 4 columns. How many elements does it have and what is its order?
(c) 8 elements, order 2 x 4 Order = rows x columns; elements = 2 x 4.
A common practice is to split data into training and testing sets in the ratio
(d) 80 : 20 70:30 is also common.
Matrix A (2 x 3) can be multiplied by matrix B only if B has
(b) 3 rows Columns of A = rows of B; result is 2 x (columns of B).
The level of measurement of 'Gender of students' is
(c) Nominal Only names/categories; no order.
Find the median of: 11, 26, 26, 35, 36, 7.
(d) 26 Sorted: 7, 11, 26, 26, 35, 36; median = 26.
The variance of a data set is 25. What is its standard deviation?
(d) 5 SD = square root of variance = sqrt(25) = 5.
Standard deviation is
(a) the square root of the variance Same unit as the data.
The level of measurement of 'Distance travelled in km' is
(d) Ratio Equal gaps and a true zero, so ratios make sense.
Ranks in a race (1st, 2nd, 3rd) are on the
(a) ordinal scale Order matters, but gaps are not equal.
Find the mean of: 39, 10, 30, 27, 27, 35.
(a) 28 Mean = 168 / 6 = 28.
Using min-max normalisation with min = 20 and max = 120, the value 60 becomes
(c) 0.4 (x - min) / (max - min) = (60 - 20) / (120 - 20) = 0.4.
Using min-max normalisation with min = 10 and max = 60, the value 50 becomes
(c) 0.8 (x - min) / (max - min) = (50 - 10) / (60 - 10) = 0.8.
The variance of a data set is 16. What is its standard deviation?
(b) 4 SD = square root of variance = sqrt(16) = 4.
The level of measurement of 'Movie rating: 1 star to 5 stars' is
(b) Ordinal Categories with an order, but gaps are not equal.
The most frequent value is the
(b) mode A set can have more than one mode.
Find the (population) variance of: 16, 10, 14, 18, 7.
(b) 16 Mean = 13. Variance = [(16 - 13)^2 + (10 - 13)^2 + (14 - 13)^2 + (18 - 13)^2 + (7 - 13)^2] / 5 = 80 / 5 = 16.
The variance of a data set is 1. What is its standard deviation?
(d) 1 SD = square root of variance = sqrt(1) = 1.
Using min-max normalisation with min = 0 and max = 50, the value 30 becomes
(b) 0.6 (x - min) / (max - min) = (30 - 0) / (50 - 0) = 0.6.
A dataset of 800 rows is split 80 : 20 into training and testing sets. How many rows are in the testing set?
(a) 160 20% of 800.
A is of order 2 x 3 and B is of order 3 x 4. What is the order of AB?
(a) 2 x 4 Columns of A = rows of B = 3, so AB is (rows of A) x (columns of B).
Find the mode of: 15, 19, 59, 15, 30, 54, 44.
(a) 15 15 occurs most often (2 times).
If A = [6 -2; 9 2], find 4A.
(c) [24 -8; 36 8] Multiply every element by 4 (scalar multiplication).
A dataset of 500 rows is split 80 : 20 into training and testing sets. How many rows are in the testing set?
(b) 100 20% of 500.
Find the (population) variance of: 8, 5, 2, 6, 4.
(b) 4 Mean = 5. Variance = [(8 - 5)^2 + (5 - 5)^2 + (2 - 5)^2 + (6 - 5)^2 + (4 - 5)^2] / 5 = 20 / 5 = 4.
Which measure is most affected by an extreme value (outlier)?
(d) Mean Mean uses every value.
Find the mean of: 59, 32, 11, 5, 26, 59, 25.
(b) 31 Mean = 217 / 7 = 31.
Which level of measurement has a true zero?
(c) Ratio 0 kg means no weight.
A is of order 3 x 2 and B is of order 2 x 2. What is the order of AB?
(b) 3 x 2 Columns of A = rows of B = 2, so AB is (rows of A) x (columns of B).
Find the mode of: 13, 5, 13, 46, 26, 42.
(b) 13 13 occurs most often (2 times).
Find the mean of: 19, 39, 35, 35, 22, 53.
(b) 33.83 Mean = 203 / 6 = 33.83.
Find the (population) variance of: 16, 10, 18, 16.
(b) 9 Mean = 15. Variance = [(16 - 15)^2 + (10 - 15)^2 + (18 - 15)^2 + (16 - 15)^2] / 4 = 36 / 4 = 9.
The variance of a data set is 36. What is its standard deviation?
(d) 6 SD = square root of variance = sqrt(36) = 6.
A matrix has 3 rows and 2 columns. How many elements does it have and what is its order?
(a) 6 elements, order 3 x 2 Order = rows x columns; elements = 3 x 2.
The level of measurement of 'Age in years' is
(b) Ratio Equal gaps and a true zero, so ratios make sense.
Data collected first-hand through surveys, interviews or experiments is
(d) primary data Secondary data is collected by someone else.
If A = [2 0; 6 -1], find 3A.
(c) [6 0; 18 -3] Multiply every element by 3 (scalar multiplication).
Find the standard deviation (population) of: 18, 18, 16, 16.
(d) 1 Mean = 17, variance = 1, SD = sqrt(1) = 1.
A = [3 6; 8 5] and B = [6 7; 9 -2]. Find the element in row 1, column 1 of AB.
(d) 72 Row 1 of A x column 1 of B = (3 x 6) + (6 x 9) = 72.
A value far away from the other values in a data set is an
(a) outlier It can distort the mean.
A = [3 0; 7 6] and B = [-3 0; 7 -2]. Find A + B.
(d) [0 0; 14 4] Add the corresponding elements.
A = [7 5; 4 1] and B = [9 5; 5 0]. Find the element in row 1, column 1 of AB.
(d) 88 Row 1 of A x column 1 of B = (7 x 9) + (5 x 5) = 88.
Converting categories like 'Red', 'Green' into numbers for a model is
(c) encoding e.g. label encoding or one-hot encoding.
A matrix with 3 rows and 2 columns has order
(d) 3 x 2 Order = rows x columns.
Find the median of: 9, 25, 46, 19, 39, 9, 58.
(a) 25 Sorted: 9, 9, 19, 25, 39, 46, 58; median = 25.
The level of measurement of 'PIN codes of cities' is
(b) Nominal Only names/categories; no order.
Temperature in degree Celsius is on the
(b) interval scale Equal gaps but no true zero (0 degree C is not 'no temperature').
A is 3 x 2 and B is 3 x 2. Is AB defined?
(a) No, columns of A (2) are not equal to rows of B (3) Multiplication needs columns of A = rows of B.
Which chart shows the relation between two numeric variables?
(c) Scatter plot Each point is (x, y).
Find the standard deviation (population) of: 14, 8, 16, 20, 17.
(d) 4 Mean = 15, variance = 16, SD = sqrt(16) = 4.
Filling missing values, removing duplicates and correcting errors is
(a) data cleaning (pre-processing) Clean data gives better models.
The level of measurement of 'Roll numbers used only as identity codes' is
(d) Nominal Only names/categories; no order.
The middle value of sorted data is the
(b) median For even count, average the two middle values.
Pearson's correlation can be calculated in MS Excel using the function
(b) =CORREL() (or =PEARSON()) Practical activity of the unit.
In MS Excel, a linear regression line can be shown on a scatter chart by adding a
(d) trendline (linear) with its equation Display equation and R-squared on chart.
For x = 3 the line y = 3x + 1 predicts 10, but the actual value is 8. What is the residual (actual - predicted)?
(c) -2 Residual = 8 - 10 = -2.
For x = 2 the line y = 2x + -3 predicts 1, but the actual value is 3. What is the residual (actual - predicted)?
(c) 2 Residual = 3 - 1 = 2.
In k-NN with k = 3, the nearest neighbours of a new point have the classes: Dog, Dog, Dog. The new point is classified as
(c) Dog Majority vote: Dog has 3 of 3 votes.
k-means (k = 2) has centroids C1 = 1 and C2 = 7. The point -2 is assigned to
(a) Cluster 1 Distance to C1 = 3, to C2 = 9; the nearer centroid wins.
The points (1, 13), (2, 18), (3, 23), (4, 28) lie exactly on a line. What is the equation of the line?
(a) y = 5x + 8 Slope = 5 (y rises 5 per unit x); at x = 0, y = 8.
A Pearson correlation coefficient of r = -0.4 shows
(b) moderate negative correlation Sign = direction; closeness to 1 = strength.
The points (1, 7), (2, 13), (3, 19), (4, 25) lie exactly on a line. What is the equation of the line?
(c) y = 6x + 1 Slope = 6 (y rises 6 per unit x); at x = 0, y = 1.
Classification predicts
(b) a category (class label) e.g. spam or not spam.
In k-NN with k = 5, the nearest neighbours of a new point have the classes: Dog, Dog, Dog, Dog, Dog. The new point is classified as
(b) Dog Majority vote: Dog has 5 of 5 votes.
A straight line passes through (5, 20) and (6, 22). What is its slope?
(d) 2 Slope = (y2 - y1) / (x2 - x1) = 2 / 1 = 2.
Classifying e-mails as spam or not spam is
(a) binary classification Two classes.
A regression line is y = 4x + 2. Predict y when x = 4.
(a) 18 y = 4 x 4 + 2 = 18.
k-NN classifies a new point by
(a) the majority class of its k nearest training points Distance-based voting.
k-means (k = 2) has centroids C1 = 1 and C2 = 7. The point 5 is assigned to
(d) Cluster 2 Distance to C1 = 4, to C2 = 2; the nearer centroid wins.
k-NN stands for
(c) k-Nearest Neighbours A simple supervised algorithm.
The value of Pearson's correlation coefficient r lies between
(c) -1 and +1 -1 perfect negative, +1 perfect positive.
Find the Euclidean distance between the points (4, 2) and (-2, 10).
(b) 10 sqrt((6)^2 + (8)^2) = sqrt(100) = 10.
k-means (k = 2) has centroids C1 = 10 and C2 = 30. The point 33 is assigned to
(d) Cluster 2 Distance to C1 = 23, to C2 = 3; the nearer centroid wins.
A straight line passes through (12, 41) and (14, 47). What is its slope?
(c) 3 Slope = (y2 - y1) / (x2 - x1) = 6 / 2 = 3.
A movie tagged as both 'comedy' and 'romance' is an example of
(a) multi-label classification One item, many labels.
Find the Euclidean distance between the points (3, 0) and (6, 4).
(d) 5 sqrt((3)^2 + (4)^2) = sqrt(25) = 5.
As the speed of a car increases, the time taken for a trip decreases. This is
(c) negative correlation One goes up, the other goes down.
In k-NN with k = 3, the nearest neighbours of a new point have the classes: Cat, Cat, Dog. The new point is classified as
(d) Cat Majority vote: Cat has 2 of 3 votes.
In the line y = mx + c, m is the
(b) slope c is the y-intercept.
r close to 0 means
(c) no linear relation Points are scattered randomly.
The distance commonly used in k-NN is
(b) Euclidean distance sqrt((x1 - x2)^2 + (y1 - y2)^2).
In k-means clustering, k is
(c) the number of clusters Chosen before running.
In k-NN with k = 5, the nearest neighbours of a new point have the classes: Cat, Dog, Cat, Cat, Cat. The new point is classified as
(d) Cat Majority vote: Cat has 4 of 5 votes.
Find the Euclidean distance between the points (1, 2) and (9, 17).
(d) 17 sqrt((8)^2 + (15)^2) = sqrt(289) = 17.
Machine learning is a branch of AI in which computers
(a) learn patterns from data and improve with experience Arthur Samuel (1959).
k-means stops when
(b) the centroids (cluster assignments) no longer change Convergence.
A Pearson correlation coefficient of r = -1.0 shows
(a) perfect negative correlation Sign = direction; closeness to 1 = strength.
k-means (k = 2) has centroids C1 = 5 and C2 = 15. The point 9 is assigned to
(d) Cluster 1 Distance to C1 = 4, to C2 = 6; the nearer centroid wins.
Find the Euclidean distance between the points (4, 3) and (3, 3).
(a) 1 sqrt((1)^2 + (0)^2) = sqrt(1) = 1.
Which type of clustering finds clusters as dense areas and can mark outliers as noise?
(d) Density-based (e.g. DBSCAN) It handles irregular shapes.
A regression line is y = 4x + 1. Predict y when x = 3.
(a) 13 y = 4 x 3 + 1 = 13.
r = +0.95 between study hours and marks shows
(d) a strong positive correlation As hours rise, marks rise.
The points (1, 9), (2, 15), (3, 21), (4, 27) lie exactly on a line. What is the equation of the line?
(c) y = 6x + 3 Slope = 6 (y rises 6 per unit x); at x = 0, y = 3.
Find the Euclidean distance between the points (6, 1) and (12, 9).
(b) 10 sqrt((6)^2 + (8)^2) = sqrt(100) = 10.
Find the Euclidean distance between the points (3, 0) and (-6, 12).
(c) 15 sqrt((9)^2 + (12)^2) = sqrt(225) = 15.
The points (1, 14), (2, 18), (3, 22), (4, 26) lie exactly on a line. What is the equation of the line?
(d) y = 4x + 10 Slope = 4 (y rises 4 per unit x); at x = 0, y = 10.
A Pearson correlation coefficient of r = 0.05 shows
(b) almost no correlation Sign = direction; closeness to 1 = strength.
For x = 2 the line y = 2x + 20 predicts 24, but the actual value is 21. What is the residual (actual - predicted)?
(a) -3 Residual = 21 - 24 = -3.
A straight line passes through (4, -3) and (8, -11). What is its slope?
(b) -2 Slope = (y2 - y1) / (x2 - x1) = -8 / 4 = -2.
A straight line passes through (8, 21) and (12, 33). What is its slope?
(c) 3 Slope = (y2 - y1) / (x2 - x1) = 12 / 4 = 3.
Grouping customers by buying behaviour for marketing is an application of
(c) clustering Customer segmentation.
In k-NN with k = 5, the nearest neighbours of a new point have the classes: Cat, Dog, Cat, Cat, Cat. The new point is classified as
(b) Cat Majority vote: Cat has 4 of 5 votes.
Classifying fruits into apple, mango and banana is
(c) multi-class classification More than two classes.
An algorithm that learns to play chess by winning (+1) or losing (-1) games uses
(c) reinforcement learning Learning from rewards.
Which type of clustering builds a tree of clusters (dendrogram)?
(a) Hierarchical clustering Agglomerative or divisive.
k-means (k = 2) has centroids C1 = 2 and C2 = 9. The point 7 is assigned to
(c) Cluster 2 Distance to C1 = 5, to C2 = 2; the nearer centroid wins.
k-means (k = 2) has centroids C1 = 10 and C2 = 30. The point 15 is assigned to
(c) Cluster 1 Distance to C1 = 5, to C2 = 15; the nearer centroid wins.
The difference between an actual value and the value predicted by the line is the
(a) residual (error) Residual = actual - predicted.
Overfitting means a model
(b) performs very well on training data but poorly on new data It memorised instead of learning patterns.
A Pearson correlation coefficient of r = -0.88 shows
(d) strong negative correlation Sign = direction; closeness to 1 = strength.
In k-means, each point is assigned to
(d) the nearest centroid Then the centroids are recomputed.
If y = 3x + 2, the predicted y for x = 4 is
(c) 14 3 x 4 + 2 = 14.
In k-NN with k = 3, the nearest neighbours of a new point have the classes: Dog, Cat, Cat. The new point is classified as
(c) Cat Majority vote: Cat has 2 of 3 votes.
A Pearson correlation coefficient of r = 0.92 shows
(c) strong positive correlation Sign = direction; closeness to 1 = strength.
k-means (k = 2) has centroids C1 = 2 and C2 = 9. The point 12 is assigned to
(d) Cluster 2 Distance to C1 = 10, to C2 = 3; the nearer centroid wins.
The three main types of machine learning are
(b) supervised, unsupervised and reinforcement Classified by the type of feedback.
In k-NN with k = 5, the nearest neighbours of a new point have the classes: Dog, Cat, Dog, Cat, Dog. The new point is classified as
(d) Dog Majority vote: Dog has 3 of 5 votes.
k-means (k = 2) has centroids C1 = 10 and C2 = 30. The point 13 is assigned to
(b) Cluster 1 Distance to C1 = 3, to C2 = 17; the nearer centroid wins.
Correlation measures
(d) the strength and direction of the linear relation between two variables Pearson's r.
A Pearson correlation coefficient of r = 1.0 shows
(d) perfect positive correlation Sign = direction; closeness to 1 = strength.
Correlation does not imply
(a) causation Ice-cream sales and drowning both rise in summer; one does not cause the other.
Data with labels is required for
(c) classification and regression Supervised algorithms need labels.
In k-NN, k is usually chosen as an odd number to
(a) avoid ties in voting (for two classes) With odd k, two classes cannot tie.
A regression line is y = 5x + 10. Predict y when x = 7.
(a) 45 y = 5 x 7 + 10 = 45.
The centre of a cluster in k-means is called the
(b) centroid It is the mean of the points in the cluster.
In linear regression, the best-fit line is the one that
(d) minimises the sum of squared errors Least squares method.
Clustering is a type of
(c) unsupervised learning It groups unlabelled data.
A regression line is y = 3x + 1. Predict y when x = 2.
(a) 7 y = 3 x 2 + 1 = 7.
In scikit-learn, LinearRegression is imported from
(a) sklearn.linear_model KNeighborsClassifier is in sklearn.neighbors; KMeans in sklearn.cluster.
Find the Euclidean distance between the points (5, 5) and (2, 9).
(a) 5 sqrt((3)^2 + (4)^2) = sqrt(25) = 5.
A cluster contains the points 11, 18, 23, 29. What is its new centroid?
(d) 20.25 Centroid = mean = 81 / 4.
A cluster contains the points 16, 22, 28. What is its new centroid?
(c) 22 Centroid = mean = 66 / 3.
The points (1, 6), (2, 9), (3, 12), (4, 15) lie exactly on a line. What is the equation of the line?
(b) y = 3x + 3 Slope = 3 (y rises 3 per unit x); at x = 0, y = 3.
Which is supervised learning?
(a) Predicting the price of a car from its age and mileage, using past sales with prices Labelled data with target values.
Regression predicts
(b) a continuous numeric value e.g. house price, temperature.
Find the Euclidean distance between the points (3, 1) and (-2, 13).
(a) 13 sqrt((5)^2 + (12)^2) = sqrt(169) = 13.
A cluster contains the points 10, 15, 17, 29. What is its new centroid?
(b) 17.75 Centroid = mean = 71 / 4.
For x = 2 the line y = -2x + 10 predicts 6, but the actual value is 10. What is the residual (actual - predicted)?
(b) 4 Residual = 10 - 6 = 4.
A model that predicts a student's marks from hours of study is
(b) a regression model The output is a number.
A Pearson correlation coefficient of r = 0.45 shows
(c) moderate positive correlation Sign = direction; closeness to 1 = strength.
A cluster contains the points 6, 10, 17, 26. What is its new centroid?
(c) 14.75 Centroid = mean = 59 / 4.
In k-NN with k = 5, the nearest neighbours of a new point have the classes: Cat, Cat, Cat, Dog, Cat. The new point is classified as
(b) Cat Majority vote: Cat has 4 of 5 votes.
A cluster contains the points 8, 16, 23. What is its new centroid?
(d) 15.67 Centroid = mean = 47 / 3.
A cluster contains the points 4, 17, 18, 29. What is its new centroid?
(b) 17 Centroid = mean = 68 / 4.
k-NN is called a 'lazy learner' because
(b) it does not build a model during training; it uses the stored data at prediction time All work happens at prediction.
IBM Project Debater is an AI system that
(b) can debate with humans on complex topics Practical activity: write an article on its facts.
Google Dialogflow, Botsify and Botpress are platforms for
(d) building chatbots Practical: ice-cream ordering chatbot.
In Dialogflow, what the user wants to do (e.g. order ice cream) is an
(b) intent Training phrases teach intents.
In Dialogflow, important details in the user's message like 'flavour = chocolate' are
(a) entities (parameters) Entities extract data.
Summarising a long article into a few lines automatically is
(c) text summarisation Extractive or abstractive.
Understanding a sentence using the sentences before it (e.g. who 'he' refers to) is
(c) discourse integration Context from previous sentences.
Human language is hard for computers mainly because of
(c) ambiguity, context, idioms and sarcasm The same words can mean different things.
Reducing 'running', 'runs', 'ran' to 'run' with a dictionary is
(a) lemmatisation Stemming just cuts endings.
The first phase of NLP, which breaks text into words and sentences, is
(a) lexical analysis Also involves tokenisation and morphology.
Natural Language Processing (NLP) enables computers to
(b) understand, interpret and generate human language Text and speech.
Which is NOT an application of NLP?
(b) Detecting a tumour in an X-ray image X-ray analysis is Computer Vision.
Finding whether a review is positive, negative or neutral is
(c) sentiment analysis Also called opinion mining.
Checking the grammar and structure of a sentence (parsing) is
(d) syntactic analysis "Agra goes to the Poonam" fails here.
Finding the dictionary meaning of a sentence is
(c) semantic analysis "Hot ice-cream" fails here.
Spam filtering of e-mails uses
(a) text classification Spam vs not spam.
A program that talks with users in natural language is a
(d) chatbot Rule-based or AI-based.
"I saw the man with the telescope" has two meanings. This is
(d) syntactic (structural) ambiguity Who has the telescope is unclear.
The word 'bat' (animal or cricket bat) shows
(a) lexical ambiguity One word, many meanings.
"Oh great, my phone died again!" is hard for NLP because of
(a) sarcasm The real sentiment is negative.
Google Translate is an application of
(c) machine translation Text from one language to another.
Sorting news articles into sports, politics and business is
(c) text classification An NLP classification problem.
Gmail suggesting the next word while you type uses
(c) next-word prediction (language models) Autocomplete is NLP.
Labelling each word as noun, verb, adjective etc. is
(b) POS (part-of-speech) tagging An advanced practical uses Python to print POS tags.
Converting written text into spoken words is
(b) text-to-speech (speech synthesis) Used in screen readers.
Breaking 'AI is fun!' into 'AI', 'is', 'fun', '!' is
(b) tokenisation Each piece is a token.
Converting speech into text, as in voice typing, is
(d) speech recognition Speech-to-text.
Detecting feelings like joy, anger or sadness in text is
(b) emotion detection More detailed than sentiment.
Stop words are
(c) very common words like 'is', 'the', 'and' often removed in text processing They carry little meaning.
A chatbot that replies only from fixed rules and keywords is a
(b) rule-based chatbot It cannot handle unexpected questions.
An AI chatbot that learns from data and understands varied questions is a
(a) smart (AI-based) chatbot e.g. ChatGPT, Gemini.
Understanding the intended meaning using real-world knowledge and situation is
(d) pragmatic analysis "Can you pass the salt?" is a request, not a question about ability.
The correct order of the phases of NLP is
(a) Lexical, Syntactic, Semantic, Discourse integration, Pragmatic From words to intended meaning.
The MIT game that asks players to decide whom a self-driving car should save is
(a) Moral Machine It explores ethical dilemmas.
'Survival of the Best Fit' is a game that teaches about
(a) bias in AI hiring systems Players see how biased data creates unfair hiring.
The video 'Humans Need Not Apply' is about
(a) automation and AI replacing human jobs Practical: summarise its insights.
A role-play on biased AI systems helps students to
(a) understand the effect of bias on different stakeholders Practical activity of the unit.
Being able to explain how an AI system reached its decision is
(d) explainability Important for loans, health, justice.
The five pillars of AI ethics (IBM) are
(a) explainability, fairness, robustness, transparency and privacy IBM's pillars of trustworthy AI.
Openly sharing how an AI system is built, what data it uses and its limits is
(c) transparency Users can trust it.
Keeping a human to review important AI decisions is called
(a) human-in-the-loop oversight It catches unfair or wrong decisions.
An AI hiring tool rejected women because past hiring data was mostly men. This shows
(d) historical bias in training data AI repeats past discrimination.
Bias that comes from the design or rules of the algorithm itself is
(d) algorithmic bias e.g. a scoring rule that unfairly weights a feature.
When an AI system makes a harmful mistake, the principle that someone must be answerable is
(b) accountability Clear responsibility for AI outcomes.
A document of rules and principles that guides how an organisation develops and uses AI is an
(c) AI policy Policies define responsible use.
Ethics is
(d) a set of moral principles about what is right and wrong AI ethics applies these to AI.
India's law that protects personal digital data (2023) is the
(d) Digital Personal Data Protection Act DPDP Act, 2023.
The European Union's law that classifies AI systems by risk level is the
(d) EU AI Act Unacceptable, high, limited and minimal risk.
Bias that comes from the beliefs and assumptions of the people who build or label the data is
(a) human (cognitive) bias e.g. stereotypes in labelling.
Deepfake videos raise ethical concerns about
(d) misinformation and consent They can deceive and harm people.
UNESCO's Recommendation on the Ethics of AI (2021) is
(c) a global standard for ethical AI adopted by member countries Adopted by 193 member states.
Bias in AI means
(d) systematic unfairness in the outputs of an AI system It can harm certain groups.
Treating all individuals and groups equally, without discrimination, is
(c) fairness Fair AI avoids biased outcomes.
NITI Aayog's approach paper for India's AI is titled
(c) Responsible AI for All India's principles for responsible AI.
Being aware that bias can exist and actively checking for it is
(a) bias awareness The first step to fairness.
Collecting only the data needed for the stated purpose follows the principle of
(d) data minimisation A privacy principle.
Which is a way to mitigate bias in AI?
(c) Using diverse, balanced data and auditing model outputs for different groups Fairness checks and human oversight.
An AI system that works reliably even with unexpected inputs or attacks is
(a) robust Robustness = reliability and security.
Protecting users' personal data and their control over it is
(a) privacy Data protection and consent.
A face-recognition system trained mostly on light-skinned faces performs poorly on dark-skinned faces. The source of bias is
(c) unrepresentative training data Data bias / sampling bias.
Bias that makes an existing unfair pattern even stronger through repeated AI decisions is
(b) bias amplification (feedback loop) e.g. predictive policing sending more police to the same areas.
The OECD AI Principles promote AI that is
(a) innovative, trustworthy and respects human rights and democratic values Adopted in 2019.
Which is an ethical concern of AI in the workplace?
(d) Job displacement and the need to reskill workers Automation can replace routine jobs.
Statement 1: The term AI was coined at the Dartmouth Conference in 1956.
Statement 2: Alan Turing proposed the Turing Test in 1950.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: All AI in use today is General AI.
Statement 2: Narrow AI is designed for a specific task.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: Deep Blue defeated Garry Kasparov in 1997.
Statement 2: AlphaGo was developed by IBM.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Reactive machines store past experiences.
Statement 2: Self-driving cars use limited memory AI.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: Supervised learning uses labelled data.
Statement 2: Reinforcement learning learns from rewards and penalties.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: Computer Vision works with text.
Statement 2: NLP works with images.
(b) Both the statements are false. Statement 1 is false; Statement 2 is false.
Statement 1: Deep learning uses neural networks with many layers.
Statement 2: Machine learning is a part of deep learning.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: AI can work without any data.
Statement 2: AI can make mistakes when it sees situations unlike its training data.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: Python is widely used in AI.
Statement 2: TensorFlow is a deep learning framework.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: Only computer engineers can have a career in AI.
Statement 2: Communication is an important skill for AI professionals.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: A data analyst prepares reports and dashboards from data.
Statement 2: A robotics engineer designs robots.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: AI is used in agriculture for crop disease detection.
Statement 2: AI is not used in banking.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Kaggle provides datasets and competitions.
Statement 2: Scikit-learn is a word processor.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: AI ethicists check that AI is fair and responsible.
Statement 2: Continuous learning is not needed in AI careers.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: A token is the smallest unit of a Python program.
Statement 2: Comments are tokens.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Tuples are immutable.
Statement 2: Lists are immutable.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Python is case-sensitive.
Statement 2: 'class' can be used as a variable name.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: break ends a loop immediately.
Statement 2: continue skips the rest of the current iteration.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: pd.read_csv() reads a CSV file into a DataFrame.
Statement 2: A Series is two-dimensional.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: fit() trains a scikit-learn model.
Statement 2: predict() gives predictions from a trained model.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: The ** operator has higher precedence than *.
Statement 2: 7 // 2 gives 3.5.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: NumPy arrays support fast numerical operations.
Statement 2: np.zeros((2, 2)) creates a 2 x 2 array of ones.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Design thinking has five stages.
Statement 2: Test is the first stage of design thinking.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: An empathy map has the quadrants Says, Thinks, Does and Feels.
Statement 2: Empathy maps are made after the product is sold.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: There are 17 Sustainable Development Goals.
Statement 2: SDG 4 is Quality Education.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: A prototype should be cheap and quick to make.
Statement 2: A prototype must be the final product.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Design thinking is iterative.
Statement 2: User feedback is not needed in design thinking.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: The 5W1H method helps decompose a problem.
Statement 2: 5W1H stands for Who, What, When, Where, Why and How.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: Blood group is nominal data.
Statement 2: Ranks in a race are ratio data.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Temperature in Celsius is interval data.
Statement 2: Weight is ratio data.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: Standard deviation is the square root of variance.
Statement 2: Variance can be negative.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: The median is affected more by outliers than the mean.
Statement 2: The mode is the most frequent value.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: Two matrices of different orders can be added.
Statement 2: The transpose of a 2 x 3 matrix is 3 x 2.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: A pie chart shows parts of a whole.
Statement 2: A line graph is best for showing trends over time.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: Normalisation scales values to a common range.
Statement 2: Data cleaning removes duplicates and fixes missing values.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: Primary data is collected first-hand.
Statement 2: Census data used from a website is primary data for you.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Regression predicts continuous values.
Statement 2: Classification predicts categories.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: Pearson's r can be 1.5.
Statement 2: r = -1 shows a perfect negative correlation.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: Correlation always means causation.
Statement 2: Two variables can be correlated by chance or a third factor.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: k-NN is a supervised learning algorithm.
Statement 2: k-means is a supervised learning algorithm.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: In k-means, k is the number of clusters.
Statement 2: In k-NN, k is the number of neighbours.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: The best-fit line minimises the sum of squared errors.
Statement 2: In y = mx + c, c is the slope.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Clustering needs labelled data.
Statement 2: Customer segmentation is an application of clustering.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: Overfitting means the model does well on new data but badly on training data.
Statement 2: A centroid is the mean of the points in a cluster.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: NLP helps computers understand human language.
Statement 2: Sarcasm is easy for computers to detect.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Lexical analysis is the first phase of NLP.
Statement 2: Pragmatic analysis comes before syntactic analysis.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Sentiment analysis finds whether text is positive, negative or neutral.
Statement 2: Machine translation converts text from one language to another.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: A rule-based chatbot can answer any question.
Statement 2: Dialogflow can be used to build chatbots.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: In Dialogflow, an intent is what the user wants to do.
Statement 2: Entities are the important details in a user message.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: Detecting a tumour in an X-ray is an NLP task.
Statement 2: Speech recognition converts speech to text.
(d) Statement 1 is false and Statement 2 is true. Statement 1 is false; Statement 2 is true.
Statement 1: Stop words are very common words with little meaning.
Statement 2: Tokenisation joins words into a paragraph.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Fairness is one of the pillars of AI ethics.
Statement 2: Transparency means hiding how AI works.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Biased training data can lead to biased AI.
Statement 2: Diverse data helps reduce bias.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: The EU AI Act classifies AI systems by risk.
Statement 2: India has no data protection law.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Moral Machine explores dilemmas of self-driving cars.
Statement 2: Survival of the Best Fit is about bias in hiring.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Statement 1: Explainability means users can understand how AI made a decision.
Statement 2: Robust AI fails easily with unexpected input.
(c) Statement 1 is true and Statement 2 is false. Statement 1 is true; Statement 2 is false.
Statement 1: Human oversight can help catch unfair AI decisions.
Statement 2: Bias awareness is the first step to reduce bias.
(a) Both the statements are true. Statement 1 is true; Statement 2 is true.
Assertion (A): All AI systems in use today are Narrow AI.
Reason (R): They are designed and trained for specific tasks.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Deep Blue is a reactive machine.
Reason (R): It decides only on the present state of the chessboard without using past experiences.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Supervised learning needs labelled data.
Reason (R): The model learns the mapping from inputs to known outputs.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Reinforcement learning is used in game-playing AI.
Reason (R): The agent learns from rewards and penalties.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Machine learning is a part of deep learning.
Reason (R): Deep learning uses neural networks with many layers.
(d) A is false but R is true. DL is a part of ML; R is true.
Assertion (A): AI lacks common sense.
Reason (R): AI learns only from the data it is trained on.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Super AI exists today.
Reason (R): Super AI would be smarter than humans in all fields.
(d) A is false but R is true. Super AI is only theoretical; R is true.
Assertion (A): Python is popular for AI careers.
Reason (R): Python has libraries such as NumPy, Pandas and scikit-learn.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Soft skills are needed in AI jobs.
Reason (R): AI professionals must explain results and work in teams.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): AI is used only in the IT industry.
Reason (R): AI is used in healthcare, finance and agriculture.
(d) A is false but R is true. AI is used across industries; R is true.
Assertion (A): Continuous learning is important for AI professionals.
Reason (R): AI tools and methods change very fast.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Data scientists need statistics.
Reason (R): Machine learning models are built on statistical ideas.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): There is a global demand for AI professionals.
Reason (R): Many industries are adopting AI.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): 'for' cannot be used as a variable name.
Reason (R): Keywords are reserved words in Python.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): A tuple's elements cannot be changed.
Reason (R): Tuples are immutable.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): print(7 // 2) gives 3.5.
Reason (R): // performs floor division.
(d) A is false but R is true. 7 // 2 = 3; R is true.
Assertion (A): Pandas is used for data analysis.
Reason (R): Pandas provides Series and DataFrame structures.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): fit() must be called before predict().
Reason (R): A model must be trained before it can make predictions.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Python is a compiled-only language.
Reason (R): Python code is executed by an interpreter.
(d) A is false but R is true. Python is interpreted; R is true.
Assertion (A): A CSV file can be opened in a spreadsheet.
Reason (R): CSV stores tabular data as plain text separated by commas.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Empathize is the first stage of design thinking.
Reason (R): Solutions should be based on a deep understanding of users.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): A prototype is tested with users.
Reason (R): Feedback helps improve the solution.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Capstone projects are linked to SDGs.
Reason (R): SDGs address real social and environmental problems.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Design thinking is a linear one-time process.
Reason (R): Design thinking allows going back to earlier stages.
(d) A is false but R is true. It is iterative; R is true.
Assertion (A): An empathy map helps understand users.
Reason (R): It records what users say, think, do and feel.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): 5W1H helps decompose a problem.
Reason (R): It asks who, what, when, where, why and how.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Temperature in Celsius is interval data.
Reason (R): It has equal intervals but no true zero.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): The mean is affected by outliers.
Reason (R): The mean uses every value in the data.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): A 2 x 3 matrix and a 3 x 2 matrix can be added.
Reason (R): Matrices of the same order can be added.
(d) A is false but R is true. Orders differ, so they cannot be added; R is true.
Assertion (A): Standard deviation is never negative.
Reason (R): It is the square root of variance, which is an average of squares.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Images can be stored as matrices.
Reason (R): Each pixel value is placed in a row and column.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Data should be cleaned before modelling.
Reason (R): Missing and wrong values reduce model accuracy.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): A histogram is used for categorical data like favourite colour.
Reason (R): A histogram shows the frequency of numeric ranges.
(d) A is false but R is true. A bar chart suits categories; R is true.
Assertion (A): Linear regression predicts continuous values.
Reason (R): It fits a straight line between input and output.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): k-means is unsupervised.
Reason (R): It groups data without using labels.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): k-NN uses distance to classify a point.
Reason (R): The nearest k points vote for the class.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Correlation proves that one variable causes another.
Reason (R): Correlation measures the linear relation between two variables.
(d) A is false but R is true. Correlation is not causation; R is true.
Assertion (A): An odd k is preferred in k-NN for two classes.
Reason (R): An odd number of votes cannot end in a tie between two classes.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): r = 0.9 shows a strong positive correlation.
Reason (R): r close to +1 means both variables increase together.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Clustering is supervised learning.
Reason (R): Clustering finds groups in unlabelled data.
(d) A is false but R is true. Clustering is unsupervised; R is true.
Assertion (A): Sarcasm is difficult for NLP.
Reason (R): The literal words and the intended meaning are different.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Pragmatic analysis is the last phase of NLP.
Reason (R): It finds the intended meaning using context and world knowledge.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): "Hot ice-cream" fails semantic analysis.
Reason (R): Semantic analysis checks the meaning of words and sentences.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): A rule-based chatbot handles every user question.
Reason (R): Rule-based chatbots reply from fixed rules.
(d) A is false but R is true. They cannot handle unexpected questions; R is true.
Assertion (A): Spam filtering is a text classification task.
Reason (R): E-mails are classified as spam or not spam.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Entities in Dialogflow store details like flavour and quantity.
Reason (R): Entities extract useful parameters from user messages.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Biased data can create a biased AI system.
Reason (R): AI learns patterns, including unfair ones, from its training data.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Explainability builds trust in AI.
Reason (R): Users can see why a decision was made.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): AI policies are not needed.
Reason (R): AI can affect privacy, jobs and fairness.
(d) A is false but R is true. Policies are needed because of these effects; R is true.
Assertion (A): Privacy is a pillar of AI ethics.
Reason (R): Users should control how their personal data is used.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Human oversight reduces harm from AI errors.
Reason (R): A human can review and correct important decisions.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Diverse training data helps reduce bias.
Reason (R): It represents different groups fairly.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Assertive communication is the most effective style at the workplace.
Reason (R): It is clear and honest while respecting others.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): "an hour" uses the article an.
Reason (R): The choice of a or an depends on the first sound of the word.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): "a university" is wrong; it should be "an university".
Reason (R): University begins with a 'yoo' consonant sound.
(d) A is false but R is true. "a university" is correct, so A is false; R is true.
Assertion (A): Sarcasm and sulking are signs of passive-aggressive communication.
Reason (R): Passive-aggressive people express anger indirectly.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): A paragraph should begin with a topic sentence.
Reason (R): The topic sentence states the main idea of the paragraph.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Grooming creates a good first impression.
Reason (R): Clean, neat and suitable appearance shows discipline and respect.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Clear goals and defined roles help in team building.
Reason (R): Every member knows what to achieve and what his or her job is.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Endless scrolling on social media falls in the 'important and urgent' box.
Reason (R): Time wasters should be eliminated.
(d) A is false but R is true. Scrolling is neither urgent nor important; A is false.
Assertion (A): Journaling helps in self-exploration.
Reason (R): Writing daily thoughts shows patterns in our behaviour.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Storming is the first stage of team development.
Reason (R): In the storming stage members have conflicts.
(d) A is false but R is true. Forming is the first stage; A is false; R is true.
Assertion (A): Ctrl + J justifies text in a word processor.
Reason (R): Justified text is aligned evenly with both margins.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): A header appears on every page of the document.
Reason (R): Header text is repeated at the top of each page.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Find and Replace can change all occurrences of a word at once.
Reason (R): Its shortcut is Ctrl + H.
(b) Both A and R are true but R is not the correct explanation of A. Both true; R gives the shortcut, not the reason.
Assertion (A): A PDF file is easily edited like a .docx file.
Reason (R): PDF keeps the same look on every device.
(d) A is false but R is true. PDF is not easily edited, so A is false; R is true.
Assertion (A): Mail merge is used to send the same letter to many people.
Reason (R): It takes names and addresses from a list.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Innovativeness is an entrepreneurial value.
Reason (R): Entrepreneurs bring new products and methods.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Entrepreneurs take reckless risks.
Reason (R): Entrepreneurs prefer moderate, calculated risks.
(d) A is false but R is true. A is false; R is true.
Assertion (A): A drone entrepreneur readily adopts new technology.
Reason (R): Drone entrepreneurs resist change even when they suffer losses.
(d) A is false but R is true. A is false; R is true.
Assertion (A): Problems around us can be sources of business ideas.
Reason (R): Solving a real problem creates value for customers.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Respect for work is an entrepreneurial value.
Reason (R): An entrepreneur believes no work is too small.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): A green economy is low-carbon and resource-efficient.
Reason (R): It aims to reduce pollution while creating jobs.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): NAPCC was launched in 2008.
Reason (R): It has eight national missions.
(b) Both A and R are true but R is not the correct explanation of A. Both true; R does not explain A.
Assertion (A): E-waste should be thrown with household garbage.
Reason (R): E-waste contains harmful metals that need safe recycling.
(d) A is false but R is true. A is false; R is true.
Assertion (A): Consumers are stakeholders in the green economy.
Reason (R): Their choice of eco-friendly products drives green markets.
(a) Both A and R are true and R is the correct explanation of A. Both statements are true, and R gives the reason why A is true.
Assertion (A): Eco-tourism damages local ecology.
Reason (R): Eco-tourism is tourism that protects nature and local culture.
(d) A is false but R is true. A is false; R is true.
Name the three methods of communication with one example each.
Verbal: spoken or written words - a phone call, an e-mail. Non-verbal: body language - a handshake, a smile. Visual: images and signs - a traffic sign, a chart.
Differentiate between assertive and aggressive communication styles.
Assertive: expresses needs clearly and calmly, listens and respects others ("I think we should..."). Aggressive: dominates, shouts, blames and ignores others' feelings ("You always spoil everything!").
What is passive-aggressive communication? Give an example.
It is expressing anger or disagreement indirectly instead of openly. Example: agreeing to do a task but deliberately doing it late, or making sarcastic comments.
Differentiate between a phrase and a sentence with examples.
A phrase is a group of words without a subject and verb that does not give complete sense: 'under the tree'. A sentence has a subject and a verb and gives complete sense: 'The children are playing under the tree.'
Name the four kinds of sentences with an example each.
Declarative: AI is useful. Interrogative: What is AI? Imperative: Open the file. Exclamatory: What a smart robot!
What are articles? When do we use 'a' and 'an'?
A, an and the are articles. 'A' is used before a consonant sound (a book, a university); 'an' before a vowel sound (an apple, an hour). 'The' is used for a particular thing.
Name the parts of a sentence. Identify them in: "Riya wrote a poem."
Subject and predicate (verb + object). Subject: Riya; verb: wrote; object: a poem.
Write any four parts of speech with one example each.
Noun (computer), pronoun (she), verb (run), adjective (smart), adverb (quickly), preposition (on), conjunction (and), interjection (wow). (Any four.)
What is a paragraph? Write its three parts.
A paragraph is a group of sentences about one main idea. Parts: topic sentence (main idea), supporting sentences (details, examples) and concluding sentence (sums up).
What is submissive communication? Why is it not effective?
In submissive communication a person avoids conflict, gives in to others and does not express own needs. It is not effective because the person's views are ignored, they feel stressed and problems are not solved.
Write two advantages of visual communication.
Visuals are understood quickly, overcome language barriers, are remembered longer and make data easy to compare (charts, graphs, signs).
Give two examples each of positive and negative body language.
Positive: smiling, eye contact, upright posture, nodding. Negative: crossed arms, looking away, slouching, fidgeting.
What is grooming? Why is it important?
Grooming is keeping ourselves clean, neat and suitably dressed. It creates a good first impression, builds confidence and shows respect for others and for the occasion.
Write any four points of a personal grooming checklist.
Bath daily; clean and ironed clothes; trimmed nails; combed hair; polished shoes; brushed teeth and fresh breath; minimal jewellery and perfume. (Any four.)
What is self-exploration? Name two techniques.
Self-exploration is understanding our own interests, values, strengths and weaknesses. Techniques: self-reflection/journal writing, taking feedback from others, personality and interest tests, trying new activities.
Differentiate between a team and a group.
A group is people who are together but may work separately. A team works together for a common goal, with shared responsibility, coordination and mutual trust.
Write any four factors that influence team building.
Clear goals, defined roles, trust, open communication, good leadership, mutual respect, recognition of effort. (Any four.)
Write two qualities of a good team member.
Listens actively, shares work and credit, respects others' views, is reliable and meets deadlines, helps resolve conflicts. (Any two.)
What is time management? Write two benefits.
Time management is planning and controlling the time spent on tasks. Benefits: less stress, more work done, better results, time for leisure.
Write any four steps of effective time management.
Set and prioritise goals; make a to-do list; create a schedule; break big tasks into smaller ones; avoid distractions; review and adjust the plan.
What is procrastination? Write two ways to overcome it.
Procrastination is putting off important tasks. Ways: break the task into small parts, set deadlines, remove distractions, start with a 5-minute rule, reward yourself after finishing.
How can a student balance study and leisure?
Make a timetable with fixed study hours and breaks, finish important tasks first, include play and hobbies, sleep well and avoid wasting time on the phone.
Explain the urgent-important (Eisenhower) matrix briefly.
Tasks are put in four boxes: important and urgent - do first; important not urgent - schedule; urgent not important - delegate; neither - delete. It helps prioritise.
Why is positive body language important in an interview?
It shows confidence, honesty and interest. A smile, firm handshake, upright posture and eye contact create a good impression.
What is a word processor? Name two examples.
A word processor is software to create, edit, format, save and print text documents. Examples: LibreOffice Writer, MS Word, Google Docs, WordPad.
Write the steps to create and save a new document in LibreOffice Writer.
Open Writer -> File > New > Text Document -> type the text -> File > Save (Ctrl + S) -> choose the folder, type the file name -> Save (.odt).
Differentiate between header and footer.
A header is text at the top margin of every page (e.g. chapter title). A footer is at the bottom margin (e.g. page numbers, date).
What is Find and Replace? Give a use.
It searches for a word or phrase and replaces it with another throughout the document. Use: changing every 'Delhi' to 'New Delhi' in a long report at once.
Name the four text alignments.
Left, centre, right and justify. Justify makes both margins straight.
What is AutoCorrect? Give an example.
AutoCorrect automatically corrects common spelling and typing mistakes as we type, e.g. 'teh' becomes 'the', '(c)' becomes the copyright symbol.
Differentiate between bulleted and numbered lists.
Bulleted lists use symbols (dots, squares) when order does not matter. Numbered lists use 1, 2, 3 or a, b, c when order matters, like steps.
Write the steps to insert a table in a Writer document.
Place the cursor -> Table > Insert Table (Ctrl + F12) -> enter the number of rows and columns -> Insert. Then type data in the cells.
What is text wrapping? Name two options.
Text wrapping decides how text flows around a picture or object. Options: wrap off, page wrap, optimal, before, after, through (in Word: square, tight, in line with text, behind text).
Name any four file formats in which a document can be saved.
.odt (Writer), .docx/.doc (Word), .pdf, .txt, .rtf, .html. (Any four.)
Write the use of:
(i) Ctrl + I makes selected text italic. (ii) Ctrl + P opens the print dialog to print the document.
Why should we use Print Preview before printing?
To check page layout, margins, page breaks and orientation, so that mistakes are corrected before printing and paper is not wasted.
What are values? Name the four entrepreneurial values.
Values are beliefs that guide behaviour and decisions. Entrepreneurial values: innovativeness, independence, outstanding performance and respect for work.
Explain innovativeness and independence as entrepreneurial values.
Innovativeness: creating new products, services or methods. Independence: preferring to make one's own decisions and be one's own boss.
What is an attitude? Give two examples of entrepreneurial attitudes.
An attitude is a way of thinking or feeling that shows in behaviour. Examples: taking moderate risks, seeking economic opportunities, using imagination and intuition, believing one can change the environment.
Why do entrepreneurs take moderate risks and not very high risks?
Very high risks may cause heavy loss, and no risk means no growth. Moderate (calculated) risks are studied and planned, so there is a good chance of success with limited loss.
Explain 'analysing situations and planning action' with an example.
Before acting, the entrepreneur studies facts and makes a plan. Example: before opening a juice stall, Ravi surveys footfall, prices of fruits and competitors, then decides the location and menu.
Explain 'respect for work' with an example.
Respect for work means valuing every type of work and doing it sincerely. Example: an owner of a restaurant who also cleans tables when needed and treats helpers with dignity.
How can a student develop entrepreneurial attitudes?
Take part in school fairs and clubs, try small projects, learn from failures, read about entrepreneurs, think of solutions to local problems, and take initiative.
Differentiate between values and attitudes.
Values are deep beliefs about what is important (honesty, respect for work). Attitudes are the ways we think and react to situations, shaped by values (positive attitude to risk).
Explain 'seeking economic opportunities' with an example.
Looking for unmet needs that can become a business. Example: noticing that people in a new housing society need laundry service and starting one.
Explain 'outstanding performance' as a value.
It means always aiming to do one's best and set high standards of quality. Entrepreneurs with this value improve products continuously and satisfy customers.
Write two differences between an entrepreneurial attitude and a non-entrepreneurial attitude.
Entrepreneurial: takes initiative, sees opportunity in problems, accepts moderate risk. Non-entrepreneurial: waits for instructions, sees only problems, avoids all risk.
Why is 'belief that you can change the environment' important for an entrepreneur?
It gives confidence to change the market, create new products and influence customers instead of accepting things as they are; it keeps the entrepreneur motivated.
What is a green economy? Name any four of its main sectors.
An economy that is low-carbon, resource-efficient and socially inclusive. Sectors: e-waste management, green transportation, renewable energy, green construction, water management. (Any four.)
What is e-waste? Why should it be managed properly?
E-waste is discarded electronic items like phones, computers and batteries. They contain toxic metals (lead, mercury, cadmium) that pollute soil and water, and valuable metals that can be recycled.
Write two features of green buildings.
Use of solar energy and natural light, rainwater harvesting, energy-efficient appliances, local and recycled materials, waste recycling. (Any two.)
What is green transportation? Give two examples.
Transport that uses less fuel and causes less pollution. Examples: electric vehicles, metro rail, CNG buses, cycling, walking, car-pooling.
Name any two policy initiatives of India for greening the economy.
National Action Plan on Climate Change (National Solar Mission etc.), Swachh Bharat Mission, FAME scheme for EVs, Ujala LED scheme, Jal Jeevan Mission, Perform Achieve and Trade scheme. (Any two.)
Who are the stakeholders in a green economy? Name any four.
Government, private companies, NGOs, local communities, consumers/citizens, financial institutions, research institutions. (Any four.)
What is the role of government in the green economy?
Makes laws and pollution standards, gives subsidies and incentives (for solar, EVs), invests in green infrastructure, spreads awareness and monitors compliance.
What is the role of private agencies in greening the economy?
Invest in renewable energy and clean technology, reduce their own emissions and waste, make eco-friendly products, follow CSR, create green jobs.
How can tourism be made green? Give two ways.
Eco-tourism that protects nature, limits on visitors in fragile areas, eco-friendly hotels using solar power and water saving, no plastic, benefits shared with local people.
How can agriculture be made green?
Organic farming, drip irrigation, crop rotation, using bio-fertilisers, not burning crop residue, solar pumps.
Write two ways of greening water management.
Rainwater harvesting, drip and sprinkler irrigation, recycling wastewater, fixing leaks, protecting lakes and rivers from pollution.
What is renewable energy? Why is it important for a green economy?
Energy from sources that are naturally replaced: sun, wind, water, biomass. It reduces pollution and greenhouse gases, does not run out and lowers dependence on fossil fuels.
Define Artificial Intelligence. Give two examples from daily life.
AI is the branch of computer science that makes machines perform tasks needing human intelligence - learning, reasoning, perception and decision making. Examples: face unlock, Google Maps traffic prediction, voice assistants, YouTube recommendations.
Write any four milestones in the evolution of AI.
1950 Turing Test (Alan Turing); 1956 term 'AI' at Dartmouth (John McCarthy); 1966 ELIZA chatbot; 1997 Deep Blue beats Kasparov; 2011 IBM Watson wins Jeopardy!; 2016 AlphaGo beats Lee Sedol; 2022 ChatGPT. (Any four.)
Differentiate between Narrow AI and General AI.
Narrow (weak) AI does one specific task well, e.g. face recognition, chess; all present AI is narrow. General (strong) AI would perform any intellectual task like a human, with understanding and reasoning across fields; it does not exist yet.
Name the four types of AI based on functionality.
Reactive machines (Deep Blue), limited memory (self-driving cars), theory of mind (understand emotions - research stage), self-aware AI (hypothetical).
Name the three domains of AI and the type of data each uses.
Data Science: numbers and tables; Computer Vision: images and videos; Natural Language Processing: text and speech.
Categorise into AI domains:
(i) NLP (ii) Computer Vision (iii) Data Science (iv) Computer Vision.
Differentiate between Machine Learning and Deep Learning.
ML learns patterns from data, often with hand-chosen features, and works with smaller data. DL is a part of ML using neural networks with many layers, learns features itself and needs large data and computing power.
Differentiate between supervised and unsupervised learning.
Supervised: trained on labelled data (input + correct output), e.g. spam detection, price prediction. Unsupervised: finds patterns in unlabelled data, e.g. customer grouping.
What is reinforcement learning? Give an example.
Learning by trial and error: an agent acts in an environment and gets rewards for good actions and penalties for bad ones, improving over time. Example: AlphaGo, a robot learning to walk.
Write two benefits and two limitations of AI.
Benefits: works 24x7, high speed and accuracy in repetitive tasks, handles big data, works in dangerous places. Limitations: lacks common sense and emotions, can be biased, costly, needs lots of data, may replace jobs.
What is cognitive computing?
Computing that simulates human thought processes - perception, learning and reasoning - to help humans make decisions, often with natural language. Example: IBM Watson.
Differentiate between features and labels with an example.
Features are the input variables; the label is the output to be predicted. House price model: area, rooms, location are features; price is the label.
Why is there a global demand for AI professionals?
Almost every industry - healthcare, finance, retail, agriculture, manufacturing - is adopting AI to save cost and improve decisions, but skilled people are few, so demand is high and growing.
Name any four job roles in AI.
Data scientist, machine learning engineer, data analyst, AI research scientist, NLP engineer, computer vision engineer, robotics engineer, AI ethicist, AI product manager. (Any four.)
Differentiate between a data analyst and a data scientist.
A data analyst studies existing data to answer business questions and makes reports and dashboards. A data scientist also builds predictive and machine learning models, needing deeper statistics and programming.
What does a machine learning engineer do?
Designs, trains, tests and deploys ML models into real products, makes them efficient and scalable, and monitors them after deployment.
Name four technical skills needed for an AI career.
Programming (Python), mathematics and statistics, machine learning, data handling (SQL, Pandas), data visualisation, deep learning frameworks, cloud tools. (Any four.)
Name four soft skills needed for an AI career.
Communication, teamwork, critical thinking, problem solving, creativity, curiosity and continuous learning, ethics. (Any four.)
Name any four tools used by AI professionals.
Python, Jupyter Notebook, NumPy, Pandas, scikit-learn, TensorFlow, PyTorch, Tableau/Power BI, Git. (Any four.)
Give two uses of AI in healthcare.
Reading X-rays, CT scans and eye scans; predicting disease risk; drug discovery; chatbots for symptoms; robotic surgery support. (Any two.)
Give two uses of AI in finance.
Fraud detection, credit scoring, algorithmic trading, chatbots for customer service, risk analysis. (Any two.)
Give two uses of AI in agriculture.
Crop disease detection from leaf images, yield and weather prediction, smart irrigation using soil sensors, drone monitoring, price forecasting. (Any two.)
How will you find ten companies hiring for AI roles? What details will you note?
Search job portals (LinkedIn, Naukri, Indeed) and company career pages for roles like 'data scientist' or 'ML engineer'. Note company name, job title, location, required qualifications, technical skills, soft skills and experience.
What is an AI ethicist? Why is this role needed?
An AI ethicist checks that AI systems are fair, transparent, private and accountable. The role is needed because AI decisions affect people's lives and can be biased or harmful.
What are tokens? Name the types of tokens in Python.
Tokens are the smallest individual units of a program. Types: keywords, identifiers, literals, operators and punctuators (delimiters).
Write any four rules for naming identifiers.
Start with a letter or underscore; can contain letters, digits and _; no spaces or special symbols; cannot be a keyword; case-sensitive.
Differentiate between interactive mode and script mode.
Interactive mode runs one statement at a time at >>> and shows the result immediately; not saved. Script mode saves the whole program as a .py file and runs it together; good for long programs.
Differentiate between a list and a tuple.
A list is mutable (can be changed) and written in [ ]. A tuple is immutable (cannot be changed) and written in ( ). Tuples are faster and safe for fixed data.
Differentiate between break and continue.
break ends the loop completely. continue skips the rest of the current iteration and goes to the next one.
Explain membership and identity operators with an example.
Membership: in, not in check presence - 3 in [1, 2, 3] is True. Identity: is, is not check whether two names refer to the same object - x is None.
What is a CSV file? Why is it used in AI?
CSV (Comma Separated Values) is a plain-text file storing tabular data with values separated by commas. It is small, simple and opened by spreadsheets and Pandas, so datasets for AI are often shared as CSV.
What is NumPy? Write two of its features.
NumPy (Numerical Python) is a library for fast arrays and mathematics. Features: n-dimensional arrays (ndarray), element-wise operations, functions like mean, sum, reshape, and much faster than lists.
Differentiate between a Pandas Series and a DataFrame.
A Series is one-dimensional labelled data (a single column). A DataFrame is two-dimensional, with rows and columns, like a table.
What is scikit-learn? Name two algorithms it provides.
Scikit-learn is a Python library for machine learning. Algorithms: LinearRegression, KNeighborsClassifier, KMeans, DecisionTreeClassifier. It also has train_test_split and accuracy_score.
Write the output: L = [10, 20, 30, 40]; print(L[1:3], L[-1], len(L))
[20, 30] 40 4
Write the output: for i in range(1, 10, 3): print(i, end=' ')
1 4 7
What is design thinking? Name its five stages.
A human-centred, iterative approach to creative problem solving. Stages: Empathize, Define, Ideate, Prototype, Test.
What is an empathy map? Name its four quadrants.
A visual tool to understand a user's experience. Quadrants: Says, Thinks, Does, Feels (with the user in the centre).
What is the 5W1H method?
A problem-decomposition method that asks Who is affected, What is the problem, When and Where does it happen, Why is it important, and How can it be solved.
Differentiate between the Ideate and Prototype stages.
Ideate: brainstorm many possible solutions without judging. Prototype: build a simple, low-cost model of the chosen idea to test it.
Why is design thinking called human-centred and iterative?
Human-centred because it starts with understanding users' needs and feelings. Iterative because after testing we go back to earlier stages to improve the solution.
What are SDGs? Name any four.
The 17 Sustainable Development Goals adopted by the UN in 2015 for 2030. E.g. No Poverty, Zero Hunger, Good Health, Quality Education, Clean Water, Climate Action. (Any four.)
What is a capstone project?
A final, team-based project that applies all the learning of the course to solve a real-world problem, usually linked to an SDG, with research, design, building and presentation.
Write the stages of the CBSE capstone project.
Prepare for the project, form a team, define the problem (aligned to an SDG), understand the users, brainstorm solutions, design the solution - and document each step.
What is a prototype? Why should it be simple?
A prototype is an early, simple model of a solution. It should be quick and cheap so that it can be tested and changed easily based on feedback.
What is problem decomposition? Give an example.
Breaking a big problem into smaller parts. Example: 'improve school canteen' -> menu planning, waiting time, waste, payments, hygiene.
Write a 'How Might We' question for students who skip breakfast.
How might we help students eat a healthy breakfast before school even when they are short of time?
What should a project abstract contain?
Title, problem statement, target users, the SDG addressed, proposed AI solution, data needed, expected impact - in a short, non-technical paragraph.
What is data literacy? Why is it important in AI?
The ability to read, understand, analyse and communicate with data. AI learns from data, so we must collect, clean, analyse and interpret it correctly to build fair and useful models.
Differentiate between primary and secondary data.
Primary data is collected first-hand for the purpose (surveys, interviews, experiments, sensors). Secondary data was collected earlier by others (census, research papers, data.gov.in, Kaggle).
Name the four levels of measurement with an example each.
Nominal: gender, blood group. Ordinal: ranks, ratings. Interval: temperature in Celsius, calendar years. Ratio: height, weight, income.
Differentiate between interval and ratio scales.
Both have equal intervals. Ratio has a true zero (0 kg = no weight), so 'twice as much' makes sense; interval has no true zero (0 degree C is not 'no temperature').
Define variance and standard deviation.
Variance = average of the squared differences from the mean: sum of (x - mean)^2 / n. Standard deviation = square root of variance; it shows the typical spread in the same unit as the data.
What is a matrix? Write two of its uses in AI.
A rectangular arrangement of numbers in rows and columns. Uses: storing images as pixel matrices, storing datasets (rows = records, columns = features), neural network weights, recommendation systems.
Write the condition for adding and for multiplying two matrices.
Addition: both matrices must have the same order. Multiplication A x B: the number of columns of A must equal the number of rows of B.
What is data pre-processing? Name two steps.
Preparing raw data for modelling. Steps: cleaning (missing values, duplicates, outliers), transformation (normalisation, encoding), reduction (removing unneeded features), integration (combining sources).
Which graph will you use:
(i) Line graph (ii) Pie chart (iii) Scatter plot (iv) Histogram.
Why do we split data into training and testing sets?
The model learns from the training set; the testing set checks how it performs on unseen data, so we know whether it will work in the real world and is not just memorising.
Differentiate between regression and classification.
Regression predicts a continuous number (price, temperature). Classification predicts a category (spam / not spam, apple / mango).
What is correlation? Name its three types.
Correlation measures the strength and direction of the linear relation between two variables. Types: positive (both increase), negative (one increases, the other decreases), zero (no relation).
What does the value of Pearson's r tell us?
Its sign gives the direction (+ positive, - negative); its size gives the strength: near 1 strong, near 0 weak or none. r lies between -1 and +1.
Write the steps of the k-NN algorithm.
Choose k; find the distance of the new point from every training point; pick the k nearest; assign the class that is most common among them (majority vote).
Write the steps of the k-means algorithm.
Choose k; pick k initial centroids; assign each point to the nearest centroid; recompute each centroid as the mean of its points; repeat until the centroids do not change.
Differentiate between k-NN and k-means.
k-NN is supervised classification; k = number of neighbours; needs labelled data. k-means is unsupervised clustering; k = number of clusters; uses unlabelled data.
Name the types of classification with an example each.
Binary: spam/not spam. Multi-class: digit 0-9. Multi-label: a movie tagged comedy and romance. (Imbalanced classification: fraud detection.)
Name the types of clustering.
Partitioning (k-means), hierarchical (agglomerative/divisive, dendrogram), density-based (DBSCAN), and grid/model-based.
What is NLP? Give two applications.
Natural Language Processing is the AI domain that lets computers understand, interpret and generate human language. Applications: chatbots, machine translation, sentiment analysis, spam filtering, voice assistants.
Why is human language complex for computers?
Words have many meanings (lexical ambiguity), sentences can be read in many ways (syntactic ambiguity), meaning depends on context, and there are idioms, sarcasm, slang and spelling mistakes.
Name the five phases of NLP in order.
Lexical analysis, syntactic analysis (parsing), semantic analysis, discourse integration, pragmatic analysis.
Differentiate between syntactic and semantic analysis with examples.
Syntactic checks grammar and structure: "School go I to" fails. Semantic checks meaning: "Colourless green ideas sleep furiously" is grammatical but meaningless.
What is sentiment analysis? Give a use.
Finding whether a text expresses a positive, negative or neutral opinion. Use: companies analyse product reviews and tweets to know customer satisfaction.
Differentiate between a rule-based chatbot and an AI chatbot.
Rule-based: replies from fixed rules and keywords; simple; fails on unexpected questions. AI chatbot: uses NLP and ML; understands varied language; learns and improves.
What are intents and entities in Dialogflow?
An intent is the purpose of the user's message (e.g. OrderIceCream). Entities are the details to extract (flavour = chocolate, size = large, quantity = 2).
What is IBM Project Debater?
An AI system by IBM that can debate complex topics with human experts: it reads millions of articles, builds arguments, listens to the opponent and gives rebuttals in natural language (first shown in 2019).
What is pragmatic analysis? Give an example.
The last phase of NLP that finds the intended meaning using context and world knowledge. "Can you open the window?" is a request, not a question about ability.
What is discourse integration?
Understanding a sentence using the sentences before it. In "Riya bought a phone. It is red.", 'it' refers to the phone.
Name the five pillars of AI ethics.
Explainability, fairness, robustness, transparency and privacy.
Differentiate between explainability and transparency.
Explainability: the AI can explain why it made a particular decision. Transparency: openness about how the system is built, what data it uses and its limits.
What is AI bias? Give an example.
Systematic unfairness in AI outputs against some group. Example: a loan AI rejecting applicants from certain areas because historical data was biased.
Name any three sources of bias in AI.
Biased or unrepresentative training data, historical bias, human (labelling) bias, algorithmic design bias, feedback loops. (Any three.)
Write two ways to mitigate bias in AI.
Use diverse and balanced data; audit outputs for different groups; involve diverse teams; keep human oversight; be transparent and allow appeals.
What is an AI policy? Why is it needed?
A set of rules and principles that guide the responsible development and use of AI. Needed to protect privacy, ensure fairness and safety, and fix accountability.
What do the games Moral Machine and Survival of the Best Fit teach?
Moral Machine: ethical dilemmas of self-driving cars and that moral choices differ among people. Survival of the Best Fit: how biased hiring data makes an AI unfair to some candidates.
Name two AI policies or laws from India or the world.
NITI Aayog's 'Responsible AI for All' (India), DPDP Act 2023 (India), EU AI Act, UNESCO Recommendation on the Ethics of AI, OECD AI Principles. (Any two.)
What is robustness in AI? Why is it important?
Robustness means an AI system works reliably and safely even with noisy, unexpected or deliberately misleading inputs. It matters in self-driving cars, medical AI and security, where a small error can cause harm.
Why is privacy an ethical issue in AI? Give two safeguards.
AI collects and analyses large amounts of personal data, which can be misused, leaked or used without consent. Safeguards: take informed consent, collect only needed data, anonymise and encrypt it, follow laws like the DPDP Act 2023.
Find the mean, variance and standard deviation of: 9, 5, 13, 17, 11.
Mean = 55 / 5 = 11
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 9 | -2 | 4 |
| 5 | -6 | 36 |
| 13 | 2 | 4 |
| 17 | 6 | 36 |
| 11 | 0 | 0 |
| Sum | 80 |
Variance = 80 / 5 = 16; SD = sqrt(16) = 4
Find the mean, variance and standard deviation of: 9, 16, 20, 18, 12.
Mean = 75 / 5 = 15
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 9 | -6 | 36 |
| 16 | 1 | 1 |
| 20 | 5 | 25 |
| 18 | 3 | 9 |
| 12 | -3 | 9 |
| Sum | 80 |
Variance = 80 / 5 = 16; SD = sqrt(16) = 4
Find the mean, variance and standard deviation of: 12, 12, 7, 2, 12.
Mean = 45 / 5 = 9
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 12 | 3 | 9 |
| 12 | 3 | 9 |
| 7 | -2 | 4 |
| 2 | -7 | 49 |
| 12 | 3 | 9 |
| Sum | 80 |
Variance = 80 / 5 = 16; SD = sqrt(16) = 4
Find the mean, variance and standard deviation of: 5, 7, 6, 9, 3.
Mean = 30 / 5 = 6
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 5 | -1 | 1 |
| 7 | 1 | 1 |
| 6 | 0 | 0 |
| 9 | 3 | 9 |
| 3 | -3 | 9 |
| Sum | 20 |
Variance = 20 / 5 = 4; SD = sqrt(4) = 2
Find the mean, variance and standard deviation of: 9, 6, 15, 7, 3.
Mean = 40 / 5 = 8
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 9 | 1 | 1 |
| 6 | -2 | 4 |
| 15 | 7 | 49 |
| 7 | -1 | 1 |
| 3 | -5 | 25 |
| Sum | 80 |
Variance = 80 / 5 = 16; SD = sqrt(16) = 4
Find the mean, variance and standard deviation of: 13, 15, 7, 11, 4.
Mean = 50 / 5 = 10
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 13 | 3 | 9 |
| 15 | 5 | 25 |
| 7 | -3 | 9 |
| 11 | 1 | 1 |
| 4 | -6 | 36 |
| Sum | 80 |
Variance = 80 / 5 = 16; SD = sqrt(16) = 4
Find the mean, variance and standard deviation of: 7, 19, 10, 11, 13.
Mean = 60 / 5 = 12
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 7 | -5 | 25 |
| 19 | 7 | 49 |
| 10 | -2 | 4 |
| 11 | -1 | 1 |
| 13 | 1 | 1 |
| Sum | 80 |
Variance = 80 / 5 = 16; SD = sqrt(16) = 4
Find the mean, variance and standard deviation of: 9, 5, 7, 8, 11.
Mean = 40 / 5 = 8
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 9 | 1 | 1 |
| 5 | -3 | 9 |
| 7 | -1 | 1 |
| 8 | 0 | 0 |
| 11 | 3 | 9 |
| Sum | 20 |
Variance = 20 / 5 = 4; SD = sqrt(4) = 2
Find the mean, variance and standard deviation of: 4, 12, 13, 10, 16.
Mean = 55 / 5 = 11
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 4 | -7 | 49 |
| 12 | 1 | 1 |
| 13 | 2 | 4 |
| 10 | -1 | 1 |
| 16 | 5 | 25 |
| Sum | 80 |
Variance = 80 / 5 = 16; SD = sqrt(16) = 4
Find the mean, variance and standard deviation of: 8, 8, 18, 8, 13.
Mean = 55 / 5 = 11
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 8 | -3 | 9 |
| 8 | -3 | 9 |
| 18 | 7 | 49 |
| 8 | -3 | 9 |
| 13 | 2 | 4 |
| Sum | 80 |
Variance = 80 / 5 = 16; SD = sqrt(16) = 4
If A = [0 0; 6 2] and B = [3 2; 5 -1], find AB.
row 1: (0)(3) + (0)(5) = 0, (0)(2) + (0)(-1) = 0 ; row 2: (6)(3) + (2)(5) = 28, (6)(2) + (2)(-1) = 10
002810
If A = [-2 1; -2 3] and B = [3 -1; 1 4], find A + B and the transpose of A.
10-17
-2-213
If A = [-2 1; 0 6] and B = [6 1; 1 6], find AB.
row 1: (-2)(6) + (1)(1) = -11, (-2)(1) + (1)(6) = 4 ; row 2: (0)(6) + (6)(1) = 6, (0)(1) + (6)(6) = 36
-114636
If A = [-2 4; 1 5] and B = [6 -1; -2 6], find A + B and the transpose of A.
43-111
-2145
If A = [0 4; 4 5] and B = [2 3; 2 3], find AB.
row 1: (0)(2) + (4)(2) = 8, (0)(3) + (4)(3) = 12 ; row 2: (4)(2) + (5)(2) = 18, (4)(3) + (5)(3) = 27
8121827
If A = [-2 6; -2 1] and B = [3 -2; -2 0], find A + B and the transpose of A.
14-41
-2-261
If A = [4 2; 3 5] and B = [-2 4; -1 1], find AB.
row 1: (4)(-2) + (2)(-1) = -10, (4)(4) + (2)(1) = 18 ; row 2: (3)(-2) + (5)(-1) = -11, (3)(4) + (5)(1) = 17
-1018-1117
If A = [6 4; 5 -1] and B = [3 5; 4 -2], find A + B and the transpose of A.
999-3
654-1
The line y = 4x + 8 fits some data.
y = 4 x 7 + 8 = 36
Residual = actual - predicted = 38 - 36 = 2
Find the Euclidean distance between P(3, 4) and Q(6, 8), and say which is nearer to P: Q or R(4, 5).
d(P, Q) = sqrt(3^2 + 4^2) = sqrt(25) = 5
d(P, R) = sqrt(1 + 1) = 1.41, so R is nearer.
k-means with k = 2, centroids C1 = 3 and C2 = 12, points 1, 4, 6, 9, 13, 15. Assign the points and find the new centroids.
Cluster 1: [1, 4, 6]; Cluster 2: [9, 13, 15]
New C1 = 3.67, new C2 = 12.33
The line y = 3x + 5 fits some data.
y = 3 x 7 + 5 = 26
Residual = actual - predicted = 28 - 26 = 2
Find the Euclidean distance between P(5, 4) and Q(11, 12), and say which is nearer to P: Q or R(6, 5).
d(P, Q) = sqrt(6^2 + 8^2) = sqrt(100) = 10
d(P, R) = sqrt(1 + 1) = 1.41, so R is nearer.
k-means with k = 2, centroids C1 = 3 and C2 = 12, points 1, 4, 6, 9, 13, 15. Assign the points and find the new centroids.
Cluster 1: [1, 4, 6]; Cluster 2: [9, 13, 15]
New C1 = 3.67, new C2 = 12.33
Write the output of the following code.
a = 10 b = 3 print(a / b, a // b, a % b, a ** 2)
3.3333333333333335 3 1 100
Each operator: / float division, // floor division, % remainder, ** power.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | a = 10 | a = 10 | |
| 2 | b = 3 | b = 3 | |
| 3 | print(a / b, a // b, a % b, a ** 2) | 3.3333333333333335 3 1 100⏎ |
Write the output of the following code.
L = [3, 8, 1, 6] L.append(9) L.sort() print(L) print(L[0] + L[-1])
[1, 3, 6, 8, 9] 10
After append and sort the list is in ascending order; first + last.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | L = [3, 8, 1, 6] | L = [3, 8, 1, 6] | |
| 2 | L.append(9) | L = [3, 8, 1, 6, 9] | |
| 3 | L.sort() | L = [1, 3, 6, 8, 9] | |
| 4 | print(L) | [1, 3, 6, 8, 9]⏎ | |
| 5 | print(L[0] + L[-1]) | 10⏎ |
Write the output of the following code.
for i in range(1, 6):
if i % 2 == 0:
print(i, 'even')
else:
print(i, 'odd')1 odd 2 even 3 odd 4 even 5 odd
The loop checks each number from 1 to 5.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | for i in range(1, 6): | i = 1 | |
| 2 | if i % 2 == 0: | condition is False | |
| 5 | print(i, 'odd') | 1 odd⏎ | |
| 1 | for i in range(1, 6): | i = 2 | |
| 2 | if i % 2 == 0: | condition is True | |
| 3 | print(i, 'even') | 2 even⏎ | |
| 1 | for i in range(1, 6): | i = 3 | |
| 2 | if i % 2 == 0: | condition is False | |
| 5 | print(i, 'odd') | 3 odd⏎ | |
| 1 | for i in range(1, 6): | i = 4 | |
| 2 | if i % 2 == 0: | condition is True | |
| 3 | print(i, 'even') | 4 even⏎ | |
| 1 | for i in range(1, 6): | i = 5 | |
| 2 | if i % 2 == 0: | condition is False | |
| 5 | print(i, 'odd') | 5 odd⏎ | |
| 1 | for i in range(1, 6): | loop ends (no more values) |
Write the output of the following code.
n = 1234
s = 0
while n > 0:
s = s + n % 10
n = n // 10
print(s)10
Each round adds the last digit and removes it.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | n = 1234 | n = 1234 | |
| 2 | s = 0 | s = 0 | |
| 3 | while n > 0: | condition is True | |
| 4 | s = s + n % 10 | s = 4 | |
| 5 | n = n // 10 | n = 123 | |
| 3 | while n > 0: | condition is True | |
| 4 | s = s + n % 10 | s = 7 | |
| 5 | n = n // 10 | n = 12 | |
| 3 | while n > 0: | condition is True | |
| 4 | s = s + n % 10 | s = 9 | |
| 5 | n = n // 10 | n = 1 | |
| 3 | while n > 0: | condition is True | |
| 4 | s = s + n % 10 | s = 10 | |
| 5 | n = n // 10 | n = 0 | |
| 3 | while n > 0: | condition is False | |
| 6 | print(s) | 10⏎ |
Write the output of the following code.
x = 5 y = 2 x = x + y y = x - y x = x - y print(x, y)
2 5
Three steps swap the values without a third variable.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | x = 5 | x = 5 | |
| 2 | y = 2 | y = 2 | |
| 3 | x = x + y | x = 7 | |
| 4 | y = x - y | y = 5 | |
| 5 | x = x - y | x = 2 | |
| 6 | print(x, y) | 2 5⏎ |
Write the output of the following code.
name = 'Python' print(name[0], name[-1], len(name)) print(name * 2)
P n 6 PythonPython
Indexing and repetition of a string.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | name = 'Python' | name = 'Python' | |
| 2 | print(name[0], name[-1], len(name)) | P n 6⏎ | |
| 3 | print(name * 2) | PythonPython⏎ |
Write the output of the following code.
L = [10, 20, 30, 40, 50] print(L[1:4]) print(L[::-1])
[20, 30, 40] [50, 40, 30, 20, 10]
Slicing: 1:4 gives indexes 1 to 3; ::-1 reverses.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | L = [10, 20, 30, 40, 50] | L = [10, 20, 30, 40, 50] | |
| 2 | print(L[1:4]) | [20, 30, 40]⏎ | |
| 3 | print(L[::-1]) | [50, 40, 30, 20, 10]⏎ |
Write the output of the following code.
total = 0
for x in [4, 7, 2, 9]:
total += x
print('Total:', total, 'Average:', total / 4)Total: 22 Average: 5.5
Running total, then average.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | total = 0 | total = 0 | |
| 2 | for x in [4, 7, 2, 9]: | x = 4 | |
| 3 | total += x | total = 4 | |
| 2 | for x in [4, 7, 2, 9]: | x = 7 | |
| 3 | total += x | total = 11 | |
| 2 | for x in [4, 7, 2, 9]: | x = 2 | |
| 3 | total += x | total = 13 | |
| 2 | for x in [4, 7, 2, 9]: | x = 9 | |
| 3 | total += x | total = 22 | |
| 2 | for x in [4, 7, 2, 9]: | loop ends (no more values) | |
| 4 | print('Total:', total, 'Average:', total / 4) | Total: 22 Average: 5.5⏎ |
Write the output of the following code.
a = int('12')
b = float('2.5')
print(a + b, type(a + b))14.5 <class 'float'>
int + float gives a float.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | a = int('12') | a = 12 | |
| 2 | b = float('2.5') | b = 2.5 | |
| 3 | print(a + b, type(a + b)) | 14.5 <class 'float'>⏎ |
Write the output of the following code.
L = [5, 10] L.extend([15, 20]) L.insert(0, 0) print(L, len(L))
[0, 5, 10, 15, 20] 5
extend adds each item, insert adds at index 0.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | L = [5, 10] | L = [5, 10] | |
| 2 | L.extend([15, 20]) | L = [5, 10, 15, 20] | |
| 3 | L.insert(0, 0) | L = [0, 5, 10, 15, 20] | |
| 4 | print(L, len(L)) | [0, 5, 10, 15, 20] 5⏎ |
Write the output of the following code.
i = 10
while i > 0:
print(i, end=' ')
i = i - 310 7 4 1
i goes 10, 7, 4, 1 and stops when it is not more than 0.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | i = 10 | i = 10 | |
| 2 | while i > 0: | condition is True | |
| 3 | print(i, end=' ') | 10 | |
| 4 | i = i - 3 | i = 7 | |
| 2 | while i > 0: | condition is True | |
| 3 | print(i, end=' ') | 7 | |
| 4 | i = i - 3 | i = 4 | |
| 2 | while i > 0: | condition is True | |
| 3 | print(i, end=' ') | 4 | |
| 4 | i = i - 3 | i = 1 | |
| 2 | while i > 0: | condition is True | |
| 3 | print(i, end=' ') | 1 | |
| 4 | i = i - 3 | i = -2 | |
| 2 | while i > 0: | condition is False |
Write the output of the following code.
marks = [45, 82, 30, 67]
for m in marks:
if m >= 33:
print(m, 'Pass')
else:
print(m, 'Fail')45 Pass 82 Pass 30 Fail 67 Pass
Each mark is checked with if-else.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | marks = [45, 82, 30, 67] | marks = [45, 82, 30, 67] | |
| 2 | for m in marks: | m = 45 | |
| 3 | if m >= 33: | condition is True | |
| 4 | print(m, 'Pass') | 45 Pass⏎ | |
| 2 | for m in marks: | m = 82 | |
| 3 | if m >= 33: | condition is True | |
| 4 | print(m, 'Pass') | 82 Pass⏎ | |
| 2 | for m in marks: | m = 30 | |
| 3 | if m >= 33: | condition is False | |
| 6 | print(m, 'Fail') | 30 Fail⏎ | |
| 2 | for m in marks: | m = 67 | |
| 3 | if m >= 33: | condition is True | |
| 4 | print(m, 'Pass') | 67 Pass⏎ | |
| 2 | for m in marks: | loop ends (no more values) |
Find the error(s) and rewrite the corrected code:
num = input('Enter a number')
if num % 2 = 0
print('Even')num = int(input('Enter a number'))
if num % 2 == 0:
print('Even')input() gives a string, so convert with int(); use == for comparison; add the colon; indent the print.
Find the error(s) and rewrite the corrected code:
for i in range(5) print(I)
for i in range(5):
print(i)Colon missing after range(5); the body must be indented; Python is case-sensitive, so I must be i.
Find the error(s) and rewrite the corrected code:
L = [1, 2, 3] L.add(4) print(l)
L = [1, 2, 3] L.append(4) print(L)
Lists use append(), not add(); the name is L, not l.
Find the error(s) and rewrite the corrected code:
Age = 15
if Age >= 18:
print('Adult')
else:
print('Minor')Age = 15
if Age >= 18:
print('Adult')
else:
print('Minor')else must line up with if, and each block must be indented the same way.
Write the output of the following code.
t = (1, 2, 3) L = list(t) L.append(4) print(t, L, len(L))
(1, 2, 3) [1, 2, 3, 4] 4
A tuple cannot change, but a list made from it can.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | t = (1, 2, 3) | t = (1, 2, 3) | |
| 2 | L = list(t) | L = [1, 2, 3] | |
| 3 | L.append(4) | L = [1, 2, 3, 4] | |
| 4 | print(t, L, len(L)) | (1, 2, 3) [1, 2, 3, 4] 4⏎ |
Write the output of the following code.
d = {'Asha': 85, 'Ravi': 72}
d['Meena'] = 91
for k in d:
print(k, d[k])Asha 85 Ravi 72 Meena 91
Adding a key, then looping over keys.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | d = {'Asha': 85, 'Ravi': 72} | d = {'Asha': 85, 'Ravi': 72} | |
| 2 | d['Meena'] = 91 | d = {'Asha': 85, 'Ravi': 72, 'Meena': 91} | |
| 3 | for k in d: | k = 'Asha' | |
| 4 | print(k, d[k]) | Asha 85⏎ | |
| 3 | for k in d: | k = 'Ravi' | |
| 4 | print(k, d[k]) | Ravi 72⏎ | |
| 3 | for k in d: | k = 'Meena' | |
| 4 | print(k, d[k]) | Meena 91⏎ | |
| 3 | for k in d: | loop ends (no more values) |
Write the output of the following code.
import numpy as np a = np.array([[1, 2], [3, 4]]) print(a.T) print(a.sum(), a.mean())
[[1 3] [2 4]] 10 2.5
Transpose and summary of a NumPy array.
Write the output of the following code.
s = 'Machine Learning'
print(s.split())
print(s.lower().count('n'))['Machine', 'Learning'] 3
split() breaks on spaces; count after lower().
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | s = 'Machine Learning' | s = 'Machine Learning' | |
| 2 | print(s.split()) | ['Machine', 'Learning']⏎ | |
| 3 | print(s.lower().count('n')) | 3⏎ |
Write the output of the following code.
n = 5
for i in range(1, n + 1):
print(' ' * (n - i) + '*' * i)* ** *** **** *****
Spaces then stars on each row: a right-aligned triangle.
Working (step by step)
| Line | Statement | Values after this step | Printed |
|---|---|---|---|
| 1 | n = 5 | n = 5 | |
| 2 | for i in range(1, n + 1): | i = 1 | |
| 3 | print(' ' * (n - i) + '*' * i) | *⏎ | |
| 2 | for i in range(1, n + 1): | i = 2 | |
| 3 | print(' ' * (n - i) + '*' * i) | **⏎ | |
| 2 | for i in range(1, n + 1): | i = 3 | |
| 3 | print(' ' * (n - i) + '*' * i) | ***⏎ | |
| 2 | for i in range(1, n + 1): | i = 4 | |
| 3 | print(' ' * (n - i) + '*' * i) | ****⏎ | |
| 2 | for i in range(1, n + 1): | i = 5 | |
| 3 | print(' ' * (n - i) + '*' * i) | *****⏎ | |
| 2 | for i in range(1, n + 1): | loop ends (no more values) |
Explain the evolution of AI with a timeline of important events.
1950: Alan Turing proposes the Turing Test. 1956: John McCarthy coins 'AI' at Dartmouth. 1966: ELIZA, the first chatbot. 1970s and late 1980s: AI winters (less funding). 1997: IBM Deep Blue beats Kasparov in chess. 2011: IBM Watson wins Jeopardy!. 2012: deep learning breakthrough in image recognition. 2016: AlphaGo beats Lee Sedol. 2022 onwards: generative AI such as ChatGPT.
Explain the types of AI based on capability and functionality.
Capability: Narrow AI - one task (all current AI); General AI - any human-level task (future); Super AI - beyond human intelligence (theoretical). Functionality: Reactive machines - no memory (Deep Blue); Limited memory - use recent data (self-driving cars); Theory of mind - understand emotions (research); Self-aware - consciousness (hypothetical).
Explain the three domains of AI with two applications each.
Data Science: works on numerical/tabular data - price prediction, fraud detection. Computer Vision: works on images and videos - face recognition, self-driving cars, medical imaging. NLP: works on text and speech - chatbots, translation, sentiment analysis.
Explain supervised, unsupervised and reinforcement learning with examples.
Supervised: learns from labelled examples to predict outputs - spam filter, house-price prediction (classification and regression). Unsupervised: finds patterns in unlabelled data - customer segmentation (clustering), market basket analysis. Reinforcement: learns by rewards and penalties through trial and error - game AI, robots.
Discuss four benefits and four limitations of AI.
Benefits: automation of repetitive tasks; 24x7 work; fast analysis of big data; fewer human errors; work in dangerous places; better medical diagnosis. Limitations: no common sense or emotions; needs lots of good data; can be biased; high cost; job displacement; privacy risks; cannot handle situations unlike its training.
Explain these AI terms: machine learning, deep learning, neural network, training data, features, label.
ML: computers learn patterns from data. DL: ML with many-layer neural networks. Neural network: layers of connected nodes inspired by the brain (input, hidden, output). Training data: examples used to teach the model. Features: input variables. Label: the output to be predicted.
Describe any four AI job roles with their responsibilities.
Data Scientist: collects and analyses data, builds predictive models. ML Engineer: builds, deploys and maintains ML models in products. Data Analyst: cleans data and makes reports and dashboards. AI Research Scientist: develops new algorithms. (Also NLP/CV engineer, robotics engineer, AI ethicist, AI product manager.)
Explain the essential technical and soft skills and tools for an AI career.
Technical: Python programming, statistics and linear algebra, machine learning, data handling (SQL, Pandas), visualisation, deep learning. Soft: communication, teamwork, critical thinking, problem solving, creativity, ethics, continuous learning. Tools: Jupyter, NumPy, Pandas, scikit-learn, TensorFlow/PyTorch, Tableau, Git, cloud platforms.
Explain the opportunities of AI in any four industries.
Healthcare: medical image analysis, disease prediction, drug discovery. Finance: fraud detection, credit scoring, chatbots. Agriculture: crop disease detection, yield and weather prediction, smart irrigation. Retail: recommendations, demand forecasting, inventory. (Also manufacturing, transport, education, entertainment.)
Compare the technical and soft skills asked by two companies for the post of 'Data Scientist'. Present it as a table description.
Company A: Python, SQL, statistics, ML (scikit-learn), Tableau; soft skills - communication, teamwork. Company B: Python, deep learning (TensorFlow), cloud (AWS), big data; soft skills - problem solving, stakeholder management. Common: Python, ML, statistics, communication. Conclusion: Python + ML + communication are core; learn cloud and DL for advanced roles.
A Class XI student likes biology and computers. Suggest an AI career path with skills to learn.
Path: AI in healthcare / bioinformatics - roles like medical data scientist or computational biologist. Skills: Python, statistics, ML, biology knowledge, medical image analysis (CV), ethics and privacy of health data. Steps: online courses (IBM SkillsBuild), Kaggle health datasets, projects, B.Tech/B.Sc. in CS or bioinformatics, internships.
Explain the five stages of design thinking with an example of improving the school library.
Empathize: interview students - books hard to find, long issue queues. Define: 'Students need a quick way to find and issue books.' Ideate: barcode scanner, app with search, self-issue kiosk, AI book recommender. Prototype: a simple app mock-up and a paper catalogue by subject. Test: students try it; feedback is used to improve.
Create an empathy map for a Class XII student preparing for board exams.
Says: "I have too much syllabus left." "I need more practice papers." Thinks: worries about results and the future; compares with friends. Does: studies late at night, uses phone for videos, skips meals. Feels: stressed, anxious, sometimes motivated. Insight: needs a study planner and stress-management support.
Apply the 5W1H method to the problem of road accidents near a school.
Who: students, parents, drivers, traffic police. What: accidents and near-misses at the school gate. When: at opening and closing times (7:30, 1:30). Where: the main road crossing in front of the school. Why: speeding vehicles, no zebra crossing, crowding. How: speed breakers, a traffic warden, AI camera to detect speeding, staggered dismissal.
Write a project abstract using the design thinking framework for an AI-based crop disease detector.
Title: Smart Leaf Doctor. Problem: small farmers lose crops because diseases are found late (SDG 2 Zero Hunger). Users: farmers with smartphones. Solution: an app that takes a photo of a leaf, uses a CV model trained on labelled leaf images to identify the disease and suggests treatment in the local language. Data: open leaf-disease image datasets and local photos. Impact: early treatment, higher yield, lower pesticide use.
What are SDGs? Explain how AI can help achieve any four SDGs.
The 17 UN goals for 2030. SDG 2 Zero Hunger: crop disease and yield prediction. SDG 3 Good Health: early disease detection from scans. SDG 4 Quality Education: personalised tutoring. SDG 13 Climate Action: predicting floods and optimising energy. (Also SDG 6 water leak detection, SDG 11 smart traffic.)
Explain the stages of a capstone project and the role of teamwork.
Stages: prepare (learn the tools), form a team with defined roles, define the problem linked to an SDG, understand users (empathy map, interviews), brainstorm solutions, design and build the solution, test it and present. Teamwork divides work by skills (research, data, coding, design, presentation), brings diverse ideas and improves quality.
Differentiate between the traditional problem-solving approach and design thinking.
Traditional: starts with the problem as given and the technology; linear; little user involvement; one solution. Design thinking: starts with users' needs (empathy); iterative; many ideas; quick prototypes tested with users; solution fits real needs.
Your school canteen wastes a lot of food. Use design thinking to propose an AI-based solution.
Empathize: talk to students and staff - portions too large, menu not liked. Define: reduce daily food waste by knowing demand. Ideate: pre-ordering app, menu voting, AI demand prediction. Prototype: a Google Form for pre-orders plus a spreadsheet model predicting demand from past days and weather. Test: run for two weeks, measure waste and improve. (SDG 12 Responsible Consumption.)
Explain the importance of empathy in solving problems with AI.
Empathy means understanding users' feelings and needs from their point of view. It ensures we solve the real problem, design solutions people will actually use, avoid bias against some users, and build trust. Without empathy, an AI may be technically good but useless or harmful.
Design thinking: explain the Ideate and Prototype stages in detail with tools used.
Ideate: generate many ideas without judging - brainstorming, mind maps, 'How Might We' questions, SCAMPER, sticky notes; then vote for the best ideas. Prototype: build a quick, cheap version - paper sketches, storyboards, mock-up screens (Canva, Figma), simple no-code AI (Teachable Machine) - to test the idea with users.
Explain problem decomposition with an example of building a chatbot for school admissions.
Break it into parts: (1) list questions parents ask (fees, documents, dates); (2) collect answers from the office; (3) design conversation flows and intents; (4) build the bot (Dialogflow); (5) test with parents; (6) deploy on the website; (7) collect feedback and update. Each part is small, clear and can be given to a team member.
Explain the five phases of NLP with an example for each.
Lexical: breaks text into tokens/words - 'AI is fun' -> AI | is | fun. Syntactic: checks grammar - 'Fun AI is' is rejected. Semantic: checks meaning - 'hot ice-cream' rejected. Discourse integration: uses previous sentences - 'Ravi came. He sat.' (He = Ravi). Pragmatic: intended meaning - 'Can you pass the salt?' is a request.
Why is human language difficult for computers? Explain with four examples.
Lexical ambiguity: 'bank' (river/money). Syntactic ambiguity: 'I saw the man with a telescope'. Context: 'It's cold' may be a request to close the window. Sarcasm/idioms: 'Great, another Monday!' or 'break a leg'. Also spelling errors, slang, multiple languages (Hinglish).
Explain how you will create an ice-cream ordering chatbot using Dialogflow.
Create an agent 'IceCreamBot'. Create intents: Welcome, OrderIceCream, OrderStatus, Goodbye. Add training phrases ('I want two chocolate cones'). Create entities: flavour (chocolate, vanilla, mango), size (small, large), quantity. Mark required parameters with prompts ('Which flavour?'). Write responses using parameters ('$quantity $flavour ice cream ordered'). Test in the simulator and integrate with a website.
Explain any four applications of NLP.
Machine translation (Google Translate). Sentiment analysis of reviews. Chatbots and voice assistants (Alexa). Spam filtering and text classification. (Also summarisation, autocomplete, speech recognition, grammar checkers.)
Explain sentiment analysis and emotion detection with examples and uses.
Sentiment analysis labels text as positive, negative or neutral: 'Battery life is amazing' -> positive. Emotion detection finds specific feelings: joy, anger, fear, sadness: 'I am furious about the delay' -> anger. Uses: brand monitoring, customer support priority, mental-health signals, election and product feedback.
Write an article on 'IBM Project Debater - interesting facts'.
IBM Project Debater is the first AI system designed to debate humans on complex topics. In 2019 it debated champion Harish Natarajan on preschool subsidies before a live audience. It searched about 400 million articles, built an opening speech in minutes, listened to the opponent and gave rebuttals. It shows advances in argument mining, NLP and speech, though the human was judged more persuasive. It is used to analyse public opinion and summarise arguments.
Differentiate between rule-based and AI-based chatbots on four points.
Working: fixed rules/keywords vs NLP + ML. Flexibility: fails on new phrasing vs understands varied questions. Learning: does not learn vs improves with data. Building: easy, cheap vs needs data and training. Examples: a menu bot on a bank site vs ChatGPT/Gemini.
Explain tokenisation, stop-word removal, stemming and lemmatisation with examples.
Tokenisation: 'I love AI!' -> I | love | AI | !. Stop-word removal: remove 'is', 'the', 'and'. Stemming: cut endings - 'studies' -> 'studi'. Lemmatisation: dictionary base word - 'studies' -> 'study', 'better' -> 'good'. These prepare text for NLP models.
Explain the challenges a chatbot faces in understanding Hinglish messages like 'Mujhe 2 chocolate ice cream chahiye'.
Mixed languages and scripts (code-mixing), Roman spelling variations (chahiye/chaiye), grammar from two languages, numbers written as digits, slang and abbreviations. Solutions: train on code-mixed data, use transliteration, add many training phrases and entity synonyms, and fall back politely when unsure.
How does a machine translation system work? Mention two difficulties.
Modern systems use neural networks trained on millions of sentence pairs in two languages; they encode the source sentence and decode it into the target language word by word using context. Difficulties: idioms ('kick the bucket'), different word order (English SVO vs Hindi SOV), ambiguous words, gender and politeness forms, rare languages with little data.
Explain the five pillars of AI ethics with an example each.
Explainability: a loan AI explains why the loan was refused. Fairness: a hiring AI treats men and women equally. Robustness: a self-driving car works safely in rain and fog. Transparency: a company discloses what data its AI uses. Privacy: a health app protects patients' records and asks consent.
What is AI bias? Explain its sources with examples.
AI bias is systematic unfairness in AI results. Sources: data bias - unrepresentative training data (face recognition weaker for dark skin); historical bias - past discrimination in data (hiring tool preferring men); human/labelling bias - stereotypes of annotators; algorithmic bias - design choices; feedback loops - predictive policing sending police to the same areas.
Suggest four strategies to mitigate bias in AI systems.
1. Collect diverse, representative data and balance classes. 2. Test and audit model results for different groups (fairness metrics). 3. Keep a human in the loop for important decisions and allow appeals. 4. Build diverse teams, document data and models (transparency), and follow ethical guidelines and policies.
Explain the need for AI policies and name four policies or guidelines.
AI affects jobs, privacy, safety and fairness, so rules are needed for responsible use, accountability and public trust. Examples: EU AI Act (risk-based), India's NITI Aayog 'Responsible AI for All', DPDP Act 2023, UNESCO Recommendation on the Ethics of AI (2021), OECD AI Principles (2019), company policies like IBM's and Google's AI principles.
Describe the Moral Machine and Survival of the Best Fit games and what they teach.
Moral Machine (MIT): the user decides whom a self-driving car with failed brakes should save - passengers or pedestrians, young or old; it shows moral dilemmas and cultural differences. Survival of the Best Fit: the user hires people quickly, then trains an AI on those choices; the AI copies the bias and rejects good candidates. Lesson: AI reflects the data and choices we give it.
Summarise the key ideas of the video 'Humans Need Not Apply'.
It argues that automation and AI can replace not only physical labour but also many white-collar jobs - drivers (self-driving cars), lawyers' paralegals, doctors' diagnosis, writers, even creative work - just as horses were replaced by engines. It warns of large-scale unemployment and says society must prepare through reskilling, new jobs and policies.
An AI used for college admissions gives lower scores to students from rural schools. Discuss the ethical issues and suggest solutions.
Issues: unfair (bias against rural students), lack of transparency and explainability, no accountability, may violate equal opportunity and increase inequality. Solutions: check the training data for representation, remove proxy features like school location, audit results by group, explain scores to applicants, keep human review and an appeal process, follow fairness policies.
Compare the AI approach of the EU AI Act and India's Responsible AI principles.
EU AI Act: a binding law; classifies AI by risk - unacceptable (banned, e.g. social scoring), high (strict rules, e.g. hiring, health), limited (transparency), minimal. India (NITI Aayog): principles of safety, equality, inclusivity, privacy, transparency, accountability and protection of human values, with the DPDP Act 2023 for data. EU is regulation-heavy; India focuses on principles and innovation.
The marks of 6 students are: 8, 35, 20, 31, 31, 19. Find:
Mean = 144 / 6 = 24
Sorted: 8, 19, 20, 31, 31, 35. Median = 25.5; Mode = 31 (2 times)
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 8 | -16 | 256 |
| 35 | 11 | 121 |
| 20 | -4 | 16 |
| 31 | 7 | 49 |
| 31 | 7 | 49 |
| 19 | -5 | 25 |
| Sum | 516 |
Variance = 516 / 6 = 86
Standard deviation = sqrt(86) = 9.27
The daily rainfall (mm) on 8 days are: 25, 28, 19, 31, 26, 38, 19, 38. Find:
Mean = 224 / 8 = 28
Sorted: 19, 19, 25, 26, 28, 31, 38, 38. Median = 27; Mode = 19, 38 (2 times)
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 25 | -3 | 9 |
| 28 | 0 | 0 |
| 19 | -9 | 81 |
| 31 | 3 | 9 |
| 26 | -2 | 4 |
| 38 | 10 | 100 |
| 19 | -9 | 81 |
| 38 | 10 | 100 |
| Sum | 384 |
Variance = 384 / 8 = 48
Standard deviation = sqrt(48) = 6.93
The runs scored in 7 matches are: 15, 29, 15, 26, 10, 8, 37. Find:
Mean = 140 / 7 = 20
Sorted: 8, 10, 15, 15, 26, 29, 37. Median = 15; Mode = 15 (2 times)
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 15 | -5 | 25 |
| 29 | 9 | 81 |
| 15 | -5 | 25 |
| 26 | 6 | 36 |
| 10 | -10 | 100 |
| 8 | -12 | 144 |
| 37 | 17 | 289 |
| Sum | 700 |
Variance = 700 / 7 = 100
Standard deviation = sqrt(100) = 10.00
The ages of 6 employees are: 6, 10, 27, 9, 10, 40. Find:
Mean = 102 / 6 = 17
Sorted: 6, 9, 10, 10, 27, 40. Median = 10; Mode = 10 (2 times)
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 6 | -11 | 121 |
| 10 | -7 | 49 |
| 27 | 10 | 100 |
| 9 | -8 | 64 |
| 10 | -7 | 49 |
| 40 | 23 | 529 |
| Sum | 912 |
Variance = 912 / 6 = 152
Standard deviation = sqrt(152) = 12.33
The marks of 8 students are: 5, 16, 6, 28, 5, 5, 24, 23. Find:
Mean = 112 / 8 = 14
Sorted: 5, 5, 5, 6, 16, 23, 24, 28. Median = 11; Mode = 5 (3 times)
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 5 | -9 | 81 |
| 16 | 2 | 4 |
| 6 | -8 | 64 |
| 28 | 14 | 196 |
| 5 | -9 | 81 |
| 5 | -9 | 81 |
| 24 | 10 | 100 |
| 23 | 9 | 81 |
| Sum | 688 |
Variance = 688 / 8 = 86
Standard deviation = sqrt(86) = 9.27
The daily rainfall (mm) on 6 days are: 18, 9, 29, 29, 36, 17. Find:
Mean = 138 / 6 = 23
Sorted: 9, 17, 18, 29, 29, 36. Median = 23.5; Mode = 29 (2 times)
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 18 | -5 | 25 |
| 9 | -14 | 196 |
| 29 | 6 | 36 |
| 29 | 6 | 36 |
| 36 | 13 | 169 |
| 17 | -6 | 36 |
| Sum | 498 |
Variance = 498 / 6 = 83
Standard deviation = sqrt(83) = 9.11
The runs scored in 8 matches are: 27, 23, 19, 14, 12, 10, 29, 10. Find:
Mean = 144 / 8 = 18
Sorted: 10, 10, 12, 14, 19, 23, 27, 29. Median = 16.5; Mode = 10 (2 times)
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 27 | 9 | 81 |
| 23 | 5 | 25 |
| 19 | 1 | 1 |
| 14 | -4 | 16 |
| 12 | -6 | 36 |
| 10 | -8 | 64 |
| 29 | 11 | 121 |
| 10 | -8 | 64 |
| Sum | 408 |
Variance = 408 / 8 = 51
Standard deviation = sqrt(51) = 7.14
The ages of 7 employees are: 27, 35, 38, 8, 27, 34, 13. Find:
Mean = 182 / 7 = 26
Sorted: 8, 13, 27, 27, 34, 35, 38. Median = 27; Mode = 27 (2 times)
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 27 | 1 | 1 |
| 35 | 9 | 81 |
| 38 | 12 | 144 |
| 8 | -18 | 324 |
| 27 | 1 | 1 |
| 34 | 8 | 64 |
| 13 | -13 | 169 |
| Sum | 784 |
Variance = 784 / 7 = 112
Standard deviation = sqrt(112) = 10.58
The marks of 7 students are: 39, 13, 25, 13, 21, 35, 29. Find:
Mean = 175 / 7 = 25
Sorted: 13, 13, 21, 25, 29, 35, 39. Median = 25; Mode = 13 (2 times)
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 39 | 14 | 196 |
| 13 | -12 | 144 |
| 25 | 0 | 0 |
| 13 | -12 | 144 |
| 21 | -4 | 16 |
| 35 | 10 | 100 |
| 29 | 4 | 16 |
| Sum | 616 |
Variance = 616 / 7 = 88
Standard deviation = sqrt(88) = 9.38
The daily rainfall (mm) on 6 days are: 15, 17, 24, 23, 17, 24. Find:
Mean = 120 / 6 = 20
Sorted: 15, 17, 17, 23, 24, 24. Median = 20; Mode = 17, 24 (2 times)
| x | x - mean | (x - mean)^2 |
|---|---|---|
| 15 | -5 | 25 |
| 17 | -3 | 9 |
| 24 | 4 | 16 |
| 23 | 3 | 9 |
| 17 | -3 | 9 |
| 24 | 4 | 16 |
| Sum | 84 |
Variance = 84 / 6 = 14
Standard deviation = sqrt(14) = 3.74
A = [2 -1; -2 3], B = [-2 -1; 2 1]. Find:
0-204
6-1-65
AB =
-6-3105
BA =
-2-121
AB is not equal to BA, so matrix multiplication is not commutative in general.
A = [1 -1; 5 4], B = [-1 1; 0 4]. Find:
0058
3-3104
AB =
-1-3-521
BA =
452016
AB is not equal to BA, so matrix multiplication is not commutative in general.
A = [5 0; 1 -1], B = [3 1; -1 3]. Find:
8102
7-13-5
AB =
1554-2
BA =
16-1-2-3
AB is not equal to BA, so matrix multiplication is not commutative in general.
A = [5 4; 3 5], B = [0 5; 4 2]. Find:
5977
10328
AB =
16332025
BA =
15252626
AB is not equal to BA, so matrix multiplication is not commutative in general.
A = [5 0; 2 0], B = [4 2; 1 -1]. Find:
923-1
11-251
AB =
201084
BA =
24030
AB is not equal to BA, so matrix multiplication is not commutative in general.
A = [-1 5; 2 -1], B = [-2 5; 1 0]. Find:
-3103-1
-1105-3
AB =
7-5-510
BA =
12-15-15
AB is not equal to BA, so matrix multiplication is not commutative in general.
Identify the level of measurement of each variable with reason: name of city, rank in class, temperature in degree C, rainfall in mm.
| Variable | Level | Reason |
|---|---|---|
| Name of city | Nominal | only labels |
| Rank in class | Ordinal | order, unequal gaps |
| Temperature (deg C) | Interval | equal gaps, no true zero |
| Rainfall (mm) | Ratio | true zero; 20 mm is twice 10 mm |
Identify the level of measurement of each variable with reason: mobile brand, T-shirt size, year of birth, pocket money.
| Variable | Level | Reason |
|---|---|---|
| Mobile brand | Nominal | only names |
| T-shirt size S/M/L | Ordinal | ordered |
| Year of birth | Interval | no true zero |
| Pocket money | Ratio | true zero |
Hours of study (x) and marks (y) of 5 students: x = [1, 2, 3, 4, 5], y = [48, 74, 74, 71, 84].
| x | y | xy | x^2 | y^2 |
|---|---|---|---|---|
| 1 | 48 | 48 | 1 | 2304 |
| 2 | 74 | 148 | 4 | 5476 |
| 3 | 74 | 222 | 9 | 5476 |
| 4 | 71 | 284 | 16 | 5041 |
| 5 | 84 | 420 | 25 | 7056 |
| Sum 15 | 351 | 1122 | 55 | 25353 |
r = [nSxy - SxSy] / sqrt([nSx^2 - (Sx)^2][nSy^2 - (Sy)^2]) = [5 x 1122 - 15 x 351] / sqrt([5 x 55 - 15^2][5 x 25353 - 351^2]) = 345 / sqrt(50 x 3564) = 0.82
The correlation is strong positive: marks tend to increase with study hours.
Hours of study (x) and marks (y) of 5 students: x = [1, 2, 3, 4, 5], y = [67, 66, 57, 69, 87].
| x | y | xy | x^2 | y^2 |
|---|---|---|---|---|
| 1 | 67 | 67 | 1 | 4489 |
| 2 | 66 | 132 | 4 | 4356 |
| 3 | 57 | 171 | 9 | 3249 |
| 4 | 69 | 276 | 16 | 4761 |
| 5 | 87 | 435 | 25 | 7569 |
| Sum 15 | 346 | 1081 | 55 | 24424 |
r = [nSxy - SxSy] / sqrt([nSx^2 - (Sx)^2][nSy^2 - (Sy)^2]) = [5 x 1081 - 15 x 346] / sqrt([5 x 55 - 15^2][5 x 24424 - 346^2]) = 215 / sqrt(50 x 2404) = 0.62
The correlation is moderate positive: marks tend to increase with study hours.
Hours of study (x) and marks (y) of 5 students: x = [1, 2, 3, 4, 5], y = [50, 65, 74, 87, 71].
| x | y | xy | x^2 | y^2 |
|---|---|---|---|---|
| 1 | 50 | 50 | 1 | 2500 |
| 2 | 65 | 130 | 4 | 4225 |
| 3 | 74 | 222 | 9 | 5476 |
| 4 | 87 | 348 | 16 | 7569 |
| 5 | 71 | 355 | 25 | 5041 |
| Sum 15 | 347 | 1105 | 55 | 24811 |
r = [nSxy - SxSy] / sqrt([nSx^2 - (Sx)^2][nSy^2 - (Sy)^2]) = [5 x 1105 - 15 x 347] / sqrt([5 x 55 - 15^2][5 x 24811 - 347^2]) = 320 / sqrt(50 x 3646) = 0.75
The correlation is strong positive: marks tend to increase with study hours.
Hours of study (x) and marks (y) of 5 students: x = [1, 2, 3, 4, 5], y = [50, 68, 70, 74, 82].
| x | y | xy | x^2 | y^2 |
|---|---|---|---|---|
| 1 | 50 | 50 | 1 | 2500 |
| 2 | 68 | 136 | 4 | 4624 |
| 3 | 70 | 210 | 9 | 4900 |
| 4 | 74 | 296 | 16 | 5476 |
| 5 | 82 | 410 | 25 | 6724 |
| Sum 15 | 344 | 1102 | 55 | 24224 |
r = [nSxy - SxSy] / sqrt([nSx^2 - (Sx)^2][nSy^2 - (Sy)^2]) = [5 x 1102 - 15 x 344] / sqrt([5 x 55 - 15^2][5 x 24224 - 344^2]) = 350 / sqrt(50 x 2784) = 0.94
The correlation is strong positive: marks tend to increase with study hours.
Hours of study (x) and marks (y) of 5 students: x = [1, 2, 3, 4, 5], y = [45, 66, 63, 83, 91].
| x | y | xy | x^2 | y^2 |
|---|---|---|---|---|
| 1 | 45 | 45 | 1 | 2025 |
| 2 | 66 | 132 | 4 | 4356 |
| 3 | 63 | 189 | 9 | 3969 |
| 4 | 83 | 332 | 16 | 6889 |
| 5 | 91 | 455 | 25 | 8281 |
| Sum 15 | 348 | 1153 | 55 | 25520 |
r = [nSxy - SxSy] / sqrt([nSx^2 - (Sx)^2][nSy^2 - (Sy)^2]) = [5 x 1153 - 15 x 348] / sqrt([5 x 55 - 15^2][5 x 25520 - 348^2]) = 545 / sqrt(50 x 6496) = 0.96
The correlation is strong positive: marks tend to increase with study hours.
Fit a linear regression line y = mx + c by the least squares method to x = [1, 4, 5, 6, 9], y = [5.5, 9, 11.5, 13, 18.5], and predict y for x = 10.
| x | y | xy | x^2 |
|---|---|---|---|
| 1 | 5.5 | 5.5 | 1 |
| 4 | 9 | 36 | 16 |
| 5 | 11.5 | 57.5 | 25 |
| 6 | 13 | 78 | 36 |
| 9 | 18.5 | 166.5 | 81 |
| Sum 25 | 57.5 | 343.5 | 159 |
m = (nSxy - SxSy) / (nSx^2 - (Sx)^2) = (5 x 343.5 - 25 x 57.5) / (5 x 159 - 25^2) = 1.65
c = mean(y) - m x mean(x) = 11.50 - 1.65 x 5.00 = 3.26. Line: y = 1.65x + 3.26
For x = 10: y = 1.65 x 10 + 3.26 = 19.74
Fit a linear regression line y = mx + c by the least squares method to x = [3, 4, 5, 6, 7], y = [12, 15, 18, 20, 22], and predict y for x = 10.
| x | y | xy | x^2 |
|---|---|---|---|
| 3 | 12 | 36 | 9 |
| 4 | 15 | 60 | 16 |
| 5 | 18 | 90 | 25 |
| 6 | 20 | 120 | 36 |
| 7 | 22 | 154 | 49 |
| Sum 25 | 87 | 460 | 135 |
m = (nSxy - SxSy) / (nSx^2 - (Sx)^2) = (5 x 460 - 25 x 87) / (5 x 135 - 25^2) = 2.50
c = mean(y) - m x mean(x) = 17.40 - 2.50 x 5.00 = 4.90. Line: y = 2.50x + 4.90
For x = 10: y = 2.50 x 10 + 4.90 = 29.90
Fit a linear regression line y = mx + c by the least squares method to x = [5, 6, 7, 8, 9], y = [15, 20, 21, 25, 27], and predict y for x = 10.
| x | y | xy | x^2 |
|---|---|---|---|
| 5 | 15 | 75 | 25 |
| 6 | 20 | 120 | 36 |
| 7 | 21 | 147 | 49 |
| 8 | 25 | 200 | 64 |
| 9 | 27 | 243 | 81 |
| Sum 35 | 108 | 785 | 255 |
m = (nSxy - SxSy) / (nSx^2 - (Sx)^2) = (5 x 785 - 35 x 108) / (5 x 255 - 35^2) = 2.90
c = mean(y) - m x mean(x) = 21.60 - 2.90 x 7.00 = 1.30. Line: y = 2.90x + 1.30
For x = 10: y = 2.90 x 10 + 1.30 = 30.30
Fit a linear regression line y = mx + c by the least squares method to x = [3, 4, 5, 6, 7], y = [7, 11, 13, 14, 15], and predict y for x = 10.
| x | y | xy | x^2 |
|---|---|---|---|
| 3 | 7 | 21 | 9 |
| 4 | 11 | 44 | 16 |
| 5 | 13 | 65 | 25 |
| 6 | 14 | 84 | 36 |
| 7 | 15 | 105 | 49 |
| Sum 25 | 60 | 319 | 135 |
m = (nSxy - SxSy) / (nSx^2 - (Sx)^2) = (5 x 319 - 25 x 60) / (5 x 135 - 25^2) = 1.90
c = mean(y) - m x mean(x) = 12.00 - 1.90 x 5.00 = 2.50. Line: y = 1.90x + 2.50
For x = 10: y = 1.90 x 10 + 2.50 = 21.50
Fit a linear regression line y = mx + c by the least squares method to x = [1, 2, 7, 8, 9], y = [6.5, 7, 16.5, 17, 17.5], and predict y for x = 10.
| x | y | xy | x^2 |
|---|---|---|---|
| 1 | 6.5 | 6.5 | 1 |
| 2 | 7 | 14 | 4 |
| 7 | 16.5 | 115.5 | 49 |
| 8 | 17 | 136 | 64 |
| 9 | 17.5 | 157.5 | 81 |
| Sum 27 | 64.5 | 429.5 | 199 |
m = (nSxy - SxSy) / (nSx^2 - (Sx)^2) = (5 x 429.5 - 27 x 64.5) / (5 x 199 - 27^2) = 1.53
c = mean(y) - m x mean(x) = 12.90 - 1.53 x 5.40 = 4.66. Line: y = 1.53x + 4.66
For x = 10: y = 1.53 x 10 + 4.66 = 19.92
Training points: (3, 1) class A; (3, 1) class A; (3, 3) class A; (6, 8) class B; (6, 6) class B; (8, 5) class B. Classify the new point (6, 3) using k-NN with k = 3 and Euclidean distance.
| Point | Class | Distance from (6, 3) |
|---|---|---|
| (3, 1) | A | sqrt(9 + 4) = 3.61 |
| (3, 1) | A | sqrt(9 + 4) = 3.61 |
| (3, 3) | A | sqrt(9 + 0) = 3.00 |
| (6, 8) | B | sqrt(0 + 25) = 5.00 |
| (6, 6) | B | sqrt(0 + 9) = 3.00 |
| (8, 5) | B | sqrt(4 + 4) = 2.83 |
3 nearest: (8, 5) B, (3, 3) A, (6, 6) B
Votes: A = 1, B = 2. The new point belongs to class B.
Training points: (2, 2) class A; (3, 2) class A; (1, 1) class A; (6, 9) class B; (8, 7) class B; (7, 7) class B. Classify the new point (3, 4) using k-NN with k = 3 and Euclidean distance.
| Point | Class | Distance from (3, 4) |
|---|---|---|
| (2, 2) | A | sqrt(1 + 4) = 2.24 |
| (3, 2) | A | sqrt(0 + 4) = 2.00 |
| (1, 1) | A | sqrt(4 + 9) = 3.61 |
| (6, 9) | B | sqrt(9 + 25) = 5.83 |
| (8, 7) | B | sqrt(25 + 9) = 5.83 |
| (7, 7) | B | sqrt(16 + 9) = 5.00 |
3 nearest: (3, 2) A, (2, 2) A, (1, 1) A
Votes: A = 3, B = 0. The new point belongs to class A.
Training points: (3, 2) class A; (1, 2) class A; (2, 1) class A; (7, 7) class B; (8, 8) class B; (7, 7) class B. Classify the new point (6, 3) using k-NN with k = 3 and Euclidean distance.
| Point | Class | Distance from (6, 3) |
|---|---|---|
| (3, 2) | A | sqrt(9 + 1) = 3.16 |
| (1, 2) | A | sqrt(25 + 1) = 5.10 |
| (2, 1) | A | sqrt(16 + 4) = 4.47 |
| (7, 7) | B | sqrt(1 + 16) = 4.12 |
| (7, 7) | B | sqrt(1 + 16) = 4.12 |
| (8, 8) | B | sqrt(4 + 25) = 5.39 |
3 nearest: (3, 2) A, (7, 7) B, (7, 7) B
Votes: A = 1, B = 2. The new point belongs to class B.
Training points: (2, 1) class A; (3, 3) class A; (2, 1) class A; (6, 5) class B; (6, 8) class B; (9, 6) class B. Classify the new point (3, 6) using k-NN with k = 3 and Euclidean distance.
| Point | Class | Distance from (3, 6) |
|---|---|---|
| (2, 1) | A | sqrt(1 + 25) = 5.10 |
| (2, 1) | A | sqrt(1 + 25) = 5.10 |
| (3, 3) | A | sqrt(0 + 9) = 3.00 |
| (6, 5) | B | sqrt(9 + 1) = 3.16 |
| (6, 8) | B | sqrt(9 + 4) = 3.61 |
| (9, 6) | B | sqrt(36 + 0) = 6.00 |
3 nearest: (3, 3) A, (6, 5) B, (6, 8) B
Votes: A = 1, B = 2. The new point belongs to class B.
Training points: (4, 4) class A; (4, 1) class A; (1, 4) class A; (9, 7) class B; (7, 9) class B; (6, 5) class B. Classify the new point (6, 5) using k-NN with k = 3 and Euclidean distance.
| Point | Class | Distance from (6, 5) |
|---|---|---|
| (4, 4) | A | sqrt(4 + 1) = 2.24 |
| (4, 1) | A | sqrt(4 + 16) = 4.47 |
| (1, 4) | A | sqrt(25 + 1) = 5.10 |
| (9, 7) | B | sqrt(9 + 4) = 3.61 |
| (7, 9) | B | sqrt(1 + 16) = 4.12 |
| (6, 5) | B | sqrt(0 + 0) = 0.00 |
3 nearest: (6, 5) B, (4, 4) A, (9, 7) B
Votes: A = 1, B = 2. The new point belongs to class B.
Apply k-means clustering (k = 2) to the points 1, 10, 19, 22, 23, 26, 27, taking the first and last points (1 and 27) as the initial centroids. Show each iteration until the clusters stop changing.
Centroids 1.00 and 27.00. Cluster 1: [1, 10]; Cluster 2: [19, 22, 23, 26, 27]. New centroids: 5.50 and 23.40
Centroids 5.50 and 23.40. Cluster 1: [1, 10]; Cluster 2: [19, 22, 23, 26, 27]. New centroids: 5.50 and 23.40
The centroids did not change, so k-means stops. Final clusters: [1, 10] and [19, 22, 23, 26, 27].
Apply k-means clustering (k = 2) to the points 4, 7, 10, 11, 21, 25, 27, taking the first and last points (4 and 27) as the initial centroids. Show each iteration until the clusters stop changing.
Centroids 4.00 and 27.00. Cluster 1: [4, 7, 10, 11]; Cluster 2: [21, 25, 27]. New centroids: 8.00 and 24.33
Centroids 8.00 and 24.33. Cluster 1: [4, 7, 10, 11]; Cluster 2: [21, 25, 27]. New centroids: 8.00 and 24.33
The centroids did not change, so k-means stops. Final clusters: [4, 7, 10, 11] and [21, 25, 27].
Apply k-means clustering (k = 2) to the points 5, 8, 10, 13, 18, 20, 29, taking the first and last points (5 and 29) as the initial centroids. Show each iteration until the clusters stop changing.
Centroids 5.00 and 29.00. Cluster 1: [5, 8, 10, 13]; Cluster 2: [18, 20, 29]. New centroids: 9.00 and 22.33
Centroids 9.00 and 22.33. Cluster 1: [5, 8, 10, 13]; Cluster 2: [18, 20, 29]. New centroids: 9.00 and 22.33
The centroids did not change, so k-means stops. Final clusters: [5, 8, 10, 13] and [18, 20, 29].
Apply k-means clustering (k = 2) to the points 4, 7, 17, 19, 23, 24, 29, taking the first and last points (4 and 29) as the initial centroids. Show each iteration until the clusters stop changing.
Centroids 4.00 and 29.00. Cluster 1: [4, 7]; Cluster 2: [17, 19, 23, 24, 29]. New centroids: 5.50 and 22.40
Centroids 5.50 and 22.40. Cluster 1: [4, 7]; Cluster 2: [17, 19, 23, 24, 29]. New centroids: 5.50 and 22.40
The centroids did not change, so k-means stops. Final clusters: [4, 7] and [17, 19, 23, 24, 29].
Apply k-means clustering (k = 2) to the points 3, 5, 12, 16, 17, 25, 26, taking the first and last points (3 and 26) as the initial centroids. Show each iteration until the clusters stop changing.
Centroids 3.00 and 26.00. Cluster 1: [3, 5, 12]; Cluster 2: [16, 17, 25, 26]. New centroids: 6.67 and 21.00
Centroids 6.67 and 21.00. Cluster 1: [3, 5, 12]; Cluster 2: [16, 17, 25, 26]. New centroids: 6.67 and 21.00
The centroids did not change, so k-means stops. Final clusters: [3, 5, 12] and [16, 17, 25, 26].
Write a program to input three numbers and print the largest using if-elif-else.
a = int(input("First: "))
b = int(input("Second: "))
c = int(input("Third: "))
if a >= b and a >= c:
big = a
elif b >= c:
big = b
else:
big = c
print("Largest =", big)
Sample run (input typed: 45, 78, 12):
First: 45 Second: 78 Third: 12 Largest = 78
Write a program to check whether a number is prime.
n = int(input("Enter a number: "))
if n < 2:
print(n, "is not prime")
else:
for i in range(2, int(n ** 0.5) + 1):
if n % i == 0:
print(n, "is not prime")
break
else:
print(n, "is prime")
Sample run (input typed: 29):
Enter a number: 29 29 is prime
Write a program to find the factorial of a number using a while loop.
n = int(input("Enter n: "))
f = 1
i = 1
while i <= n:
f = f * i
i = i + 1
print("Factorial of", n, "=", f)
Sample run (input typed: 6):
Enter n: 6 Factorial of 6 = 720
Write a program to print the first n terms of the Fibonacci series.
n = int(input("How many terms? "))
a, b = 0, 1
for i in range(n):
print(a, end=" ")
a, b = b, a + b
print()
Sample run (input typed: 10):
How many terms? 10 0 1 1 2 3 5 8 13 21 34
Write a program to check whether a string is a palindrome.
s = input("Enter a word: ")
if s.lower() == s.lower()[::-1]:
print(s, "is a palindrome")
else:
print(s, "is not a palindrome")
Sample run (input typed: Madam):
Enter a word: Madam Madam is a palindrome
Write a program to count vowels, consonants, digits and spaces in a string.
s = input("Enter a sentence: ")
v = c = d = sp = 0
for ch in s:
if ch.lower() in "aeiou":
v += 1
elif ch.isalpha():
c += 1
elif ch.isdigit():
d += 1
elif ch == " ":
sp += 1
print("Vowels:", v, "Consonants:", c, "Digits:", d, "Spaces:", sp)
Sample run (input typed: AI 843 is fun):
Enter a sentence: AI 843 is fun Vowels: 4 Consonants: 3 Digits: 3 Spaces: 3
Write a program to find the sum of digits and the reverse of a number.
n = int(input("Enter a number: "))
s = 0
rev = 0
while n > 0:
d = n % 10
s += d
rev = rev * 10 + d
n //= 10
print("Sum of digits =", s)
print("Reverse =", rev)
Sample run (input typed: 4527):
Enter a number: 4527 Sum of digits = 18 Reverse = 7254
Write a program to store marks of 5 students in a list and print the highest, lowest, average and the marks above average.
marks = [78, 92, 65, 88, 71]
avg = sum(marks) / len(marks)
print("Highest:", max(marks), " Lowest:", min(marks))
print("Average:", avg)
print("Above average:", [m for m in marks if m > avg])
Sample run:
Highest: 92 Lowest: 65 Average: 78.8 Above average: [92, 88]
Write a program to create a dictionary of 4 students and their marks, and print the names of students scoring more than 75.
d = {"Asha": 82, "Ravi": 67, "Meena": 91, "John": 74}
for name in d:
if d[name] > 75:
print(name, d[name])
Sample run:
Asha 82 Meena 91
Write a program to compute the mean, median, mode, variance and standard deviation of a list using the statistics module.
import statistics as st
data = [12, 15, 15, 18, 20, 22, 25]
print("Mean :", round(st.mean(data), 2))
print("Median :", st.median(data))
print("Mode :", st.mode(data))
print("Variance:", round(st.pvariance(data), 2))
print("Std dev :", round(st.pstdev(data), 2))
Sample run:
Mean : 18.14 Median : 18 Mode : 15 Variance: 17.55 Std dev : 4.19
Write a program using NumPy to create an array of marks and print its shape, mean, maximum, and the marks above 60.
import numpy as np
marks = np.array([45, 78, 62, 90, 55, 81])
print("Shape:", marks.shape)
print("Mean :", marks.mean())
print("Max :", marks.max())
print("Above 60:", marks[marks > 60])
Sample run:
Shape: (6,) Mean : 68.5 Max : 90 Above 60: [78 62 90 81]
Write a program using NumPy to add and multiply two 2 x 2 matrices and find the transpose.
import numpy as np
A = np.array([[1, 2], [3, 4]])
B = np.array([[5, 6], [7, 8]])
print("A + B =\n", A + B)
print("A x B =\n", np.dot(A, B))
print("Transpose of A =\n", A.T)
Sample run:
A + B = [[ 6 8] [10 12]] A x B = [[19 22] [43 50]] Transpose of A = [[1 3] [2 4]]
Write a program using Pandas to create a DataFrame of 4 students with Name and Marks, add a Result column (Pass if marks >= 33) and print it.
import pandas as pd
df = pd.DataFrame({"Name": ["Asha", "Ravi", "Meena", "John"], "Marks": [82, 29, 91, 45]})
df["Result"] = ["Pass" if m >= 33 else "Fail" for m in df["Marks"]]
print(df)
print("Average marks:", df["Marks"].mean())
Sample run:
Name Marks Result 0 Asha 82 Pass 1 Ravi 29 Fail 2 Meena 91 Pass 3 John 45 Pass Average marks: 61.75
Write a program using Pandas to create a Series of monthly rainfall and print the months with rainfall above 100 mm and the total.
import pandas as pd
rain = pd.Series([20, 35, 60, 120, 210, 95], index=["Jan", "Feb", "Mar", "Apr", "May", "Jun"])
print(rain[rain > 100])
print("Total rainfall:", rain.sum())
Sample run:
Apr 120 May 210 dtype: int64 Total rainfall: 540
Write a program to calculate Pearson's correlation coefficient between hours studied and marks (without libraries).
x = [1, 2, 3, 4, 5]
y = [52, 58, 65, 70, 80]
n = len(x)
sx, sy = sum(x), sum(y)
sxy = sum(a * b for a, b in zip(x, y))
sx2 = sum(a * a for a in x)
sy2 = sum(b * b for b in y)
r = (n * sxy - sx * sy) / ((n * sx2 - sx ** 2) * (n * sy2 - sy ** 2)) ** 0.5
print("r =", round(r, 3))
Sample run:
r = 0.994
Write a program to find the slope and intercept of the best-fit line y = mx + c by the least squares method and predict y for x = 6.
x = [1, 2, 3, 4, 5]
y = [3, 5, 7, 9, 11]
n = len(x)
m = (n * sum(a * b for a, b in zip(x, y)) - sum(x) * sum(y)) / (n * sum(a * a for a in x) - sum(x) ** 2)
c = sum(y) / n - m * sum(x) / n
print("Slope m =", m, " Intercept c =", c)
print("Predicted y for x = 6:", m * 6 + c)
Sample run:
Slope m = 2.0 Intercept c = 1.0 Predicted y for x = 6: 13.0
Write a program to classify a new point using k-NN (k = 3) with Euclidean distance (without libraries).
train = [((1, 2), "A"), ((2, 3), "A"), ((3, 1), "A"), ((6, 5), "B"), ((7, 7), "B"), ((8, 6), "B")]
new = (3, 3)
d = []
for (x, y), lab in train:
dist = ((x - new[0]) ** 2 + (y - new[1]) ** 2) ** 0.5
d.append((round(dist, 2), lab))
d.sort()
print("Nearest 3:", d[:3])
votes = [lab for _, lab in d[:3]]
print("Class:", max(set(votes), key=votes.count))
Sample run:
Nearest 3: [(1.0, 'A'), (2.0, 'A'), (2.24, 'A')] Class: A
Write a program to perform one iteration of k-means (k = 2) on 1-D data with initial centroids 2 and 10.
data = [1, 2, 3, 8, 9, 10, 11]
c1, c2 = 2, 10
g1 = [x for x in data if abs(x - c1) <= abs(x - c2)]
g2 = [x for x in data if abs(x - c1) > abs(x - c2)]
print("Cluster 1:", g1, " Cluster 2:", g2)
print("New centroids:", sum(g1) / len(g1), sum(g2) / len(g2))
Sample run:
Cluster 1: [1, 2, 3] Cluster 2: [8, 9, 10, 11] New centroids: 2.0 9.5
Write a program to read a CSV file of students and print the names and marks of those who scored above 80.
import csv
with open("marks.csv", "w", newline="") as f:
w = csv.writer(f)
w.writerows([["Name", "Marks"], ["Asha", 85], ["Ravi", 72], ["Meena", 91]])
with open("marks.csv") as f:
r = csv.reader(f)
next(r)
for row in r:
if int(row[1]) > 80:
print(row[0], row[1])
Sample run:
Asha 85 Meena 91
Write a simple rule-based chatbot in Python that answers greetings, timings and fees, and stops on 'bye'.
replies = {"hi": "Hello! How can I help you?", "timing": "School timing is 7:30 am to 1:30 pm.", "fees": "Please visit the office for the fee details.", "bye": "Goodbye!"}
for msg in ["hi", "what is the timing", "fees please", "bye"]:
print("You:", msg)
ans = "Sorry, I did not understand."
for key in replies:
if key in msg.lower():
ans = replies[key]
print("Bot:", ans)
Sample run:
You: hi Bot: Hello! How can I help you? You: what is the timing Bot: School timing is 7:30 am to 1:30 pm. You: fees please Bot: Please visit the office for the fee details. You: bye Bot: Goodbye!
Write a program to normalise a list of values to the range 0 to 1 using min-max normalisation.
data = [20, 35, 50, 65, 80]
lo, hi = min(data), max(data)
norm = [round((x - lo) / (hi - lo), 2) for x in data]
print("Normalised:", norm)
Sample run:
Normalised: [0.0, 0.25, 0.5, 0.75, 1.0]
Write a program to draw a bar chart of the number of students in each house using matplotlib.
import matplotlib.pyplot as plt
houses = ["Raman", "Shivaji", "Tagore", "Ashoka"]
count = [42, 38, 45, 40]
plt.bar(houses, count, color="teal")
plt.title("Students in each house")
plt.xlabel("House")
plt.ylabel("Students")
plt.show()
Sample run:
Write a program to draw a line graph of monthly rainfall using matplotlib.
import matplotlib.pyplot as plt
months = ["Jan", "Feb", "Mar", "Apr", "May", "Jun"]
rain = [20, 35, 60, 120, 210, 95]
plt.plot(months, rain, marker="o", color="blue")
plt.title("Monthly rainfall")
plt.ylabel("Rainfall (mm)")
plt.show()
Sample run:
Explain the elements of non-verbal communication with examples.
Facial expressions - a smile shows happiness; gestures - thumbs-up; posture - upright shows confidence; eye contact - shows honesty and interest; touch - a handshake; space - personal distance in conversation; paralanguage - tone, pitch, speed of voice; appearance - dress and grooming.
Explain the four communication styles with examples.
Assertive - clear and respectful ('Please mention my part in the report'); Aggressive - loud and blaming ('You are a cheat!'); Passive-aggressive - indirect anger, sarcasm ('Congratulations on your great idea'); Submissive - gives in and stays silent. Assertive is the best.
Write the rules for using the articles a, an and the with examples.
'a' is used before a consonant sound - a book, a university, a one-rupee coin. 'an' is used before a vowel sound - an apple, an hour, an MBA. 'the' is used for a particular or unique thing - the Sun, the Ganga, the report you sent. The choice depends on sound, not spelling.
Explain the structure of a paragraph with an example.
A paragraph has a topic sentence (main idea), supporting sentences (facts, examples, reasons) and a concluding sentence. Example: 'Regular exercise improves performance (topic). It increases blood flow to the brain and reduces stress (support). Therefore schools should include daily activity (conclusion).'
Identify the parts of speech in: "Oh! She quickly solved the hard puzzle and smiled."
Oh - interjection; She - pronoun; quickly - adverb; solved - verb; the - article (determiner); hard - adjective; puzzle - noun; and - conjunction; smiled - verb.
Prepare a personal grooming checklist for an interview.
Daily bath and clean teeth; neat, trimmed hair; short clean nails; clean, ironed formal clothes; polished formal shoes and matching socks; minimal accessories; documents in a file; upright posture, smile and eye contact.
Explain any four methods of self-exploration.
Reflection and journaling - shows patterns in behaviour; feedback from teachers, parents and friends; interest, aptitude and personality tests; trying new activities such as clubs and volunteering; SWOT analysis of strengths, weaknesses, opportunities and threats.
Explain the factors of team building and teamwork.
Team building factors: clear goal, defined roles, trust, open communication, good leadership and recognition. Teamwork factors: cooperation, respect for others, sharing credit, accountability and peaceful conflict resolution.
Explain the five stages of team development.
Forming - members meet and get to know each other; Storming - conflicts and competition arise; Norming - rules, roles and trust develop; Performing - the team works efficiently towards the goal; Adjourning - the task ends and the team separates.
Explain the Eisenhower (urgent-important) matrix and any two time-management tips.
Important and urgent - do first; important but not urgent - schedule; urgent but not important - delegate or minimise; neither - eliminate. Tips: make a to-do list and schedule; break big tasks; use the Pomodoro method; avoid procrastination.
Explain any four text-formatting options in a word processor with shortcuts.
Bold - Ctrl + B; Italic - Ctrl + I; Underline - Ctrl + U; font face, size and colour from the formatting toolbar; superscript (Ctrl + Shift + P) and subscript (Ctrl + Shift + B) in Writer; highlight for marking text.
Write the steps to insert a table and a picture in LibreOffice Writer.
Table: Table > Insert Table (Ctrl + F12), choose rows and columns, click Insert; type data; merge cells or add borders as needed. Picture: Insert > Image, choose the file, click Open; right-click > Wrap to set text wrapping; resize using the handles.
What is a header and footer? How are page numbers inserted?
A header is text repeated at the top of every page and a footer at the bottom - e.g. school name, title, page number. Insert > Header and Footer > Header/Footer > Default. Page number: click in the footer, then Insert > Field > Page Number.
Explain spelling and grammar check, AutoCorrect, thesaurus and word count.
Spelling and grammar (F7) finds and suggests corrections; AutoCorrect fixes common typing errors automatically (teh -> the); Thesaurus (Ctrl + F7) gives synonyms; Word Count (Tools > Word Count) shows the number of words and characters.
Differentiate between Save and Save As. Name four file formats a document can be saved in.
Save (Ctrl + S) stores changes in the same file; Save As (Ctrl + Shift + S) saves a copy with a new name, place or format. Formats: .odt, .docx, .pdf (export), .txt, .rtf.
Explain any four entrepreneurial values with examples.
Innovativeness - eco-friendly packaging; Independence - takes own decisions; Outstanding performance - high quality standards; Respect for work - no work is small; Honesty - fair dealing with customers; Social responsibility - caring for society and environment. (Any four.)
Explain the types of entrepreneurs: innovative, imitative, Fabian, drone and social.
Innovative - introduces new ideas and technology; Imitative - copies successful ideas and adapts them; Fabian - cautious, changes only when necessary; Drone - resists change even with losses; Social - solves a social problem while earning.
Explain the steps of generating a business idea.
Observe problems around you; brainstorm many ideas; screen ideas for practicality; talk to possible customers; choose the best idea and test it on a small scale. Sources: hobbies and skills, problems of people, trends, government schemes and customer feedback.
Explain any four barriers to entrepreneurship and their solutions.
Lack of money - savings, bank and Mudra loans; fear of failure - start small and learn; lack of knowledge - training and mentors; competition - better quality and service; legal formalities - expert help and Udyam registration.
Write any four entrepreneurial competencies and explain them.
Decisiveness - takes quick, clear decisions; Initiative - starts things without waiting; Interpersonal skills - good relations with customers and staff; Perseverance - continues despite failure; Organisational skills - plans and uses resources well; Problem solving; Financial literacy.
What is a green economy? Differentiate it from a brown economy.
A green economy is low-carbon, resource-efficient and socially inclusive. A brown economy depends on coal and oil, causes high pollution and waste and uses resources faster than nature can renew them; a green economy uses renewable energy, low emissions, recycling and fair use of resources.
Explain any four sectors of the green economy with green practices.
Renewable energy - rooftop solar, wind farms; green transport - metro, EVs, cycling; green construction - GRIHA buildings with natural light; water management - rainwater harvesting, drip irrigation; e-waste management - safe recycling; sustainable agriculture - organic farming.
Name the eight missions of the National Action Plan on Climate Change.
National Solar Mission; Enhanced Energy Efficiency; Sustainable Habitat; Water; Sustaining the Himalayan Ecosystem; Green India; Sustainable Agriculture; Strategic Knowledge for Climate Change.
Explain the role of stakeholders in the green economy.
Government - policies, laws and subsidies; private companies - clean technology and less waste; NGOs - awareness and community projects; communities - protecting local resources; consumers - eco-friendly choices; banks - green finance; research bodies - new clean technology.
Explain greening of any four sectors.
Cities - green spaces, public transport, waste plants; tourism - eco-tourism and no plastic; industry - cleaner production and effluent treatment; forests - afforestation; fisheries - no overfishing; agriculture - natural farming and solar pumps.
Tap an option to check your answer at once; your score is saved
Ten multiple-choice questions picked at random. Tap an option to answer. You get your score at the end.
Ten ready papers with blueprints and a random paper maker, on the CBSE pattern for Class XI (843)
Each paper has 21 questions for 50 marks and 2 hours, on the design of the CBSE Sample Question Paper (Class XII 843 pattern, used by schools for Class XI): any 4 of 6 in Q1, any 5 of 6 in Q2 to Q5, any 3 of Q6 to Q10, any 4 of Q11 to Q16 and any 3 of Q17 to Q21. Every paper has Python output, statistics, matrix and ML numericals in Section A, and a Python program, a design thinking case, a statistics or matrix numerical, an ML numerical and an NLP case in Q17 to Q21. No question is repeated: the ten papers use 460 different questions.
Downloads and printing work when this page is opened as an HTML file in a browser.
Choose Paper 1 to Paper 10, or make a random paper. The blueprint and the paper open here.
Read this the evening before and again at the exam centre gate