PART B ▪ UNIT 3 · Math for AI

  root@vm-learning
  ~
  $
  open
  ch-b3
  

PART B ▪ UNIT 3

Math for AI

Statistics & Probability – The Language of AI

Mathematics is the language of AI. Just as humans use language to express thoughts, AI uses math to understand data and take decisions. Every AI system – from Google Assistant to self-driving cars – works on mathematical calculations.

You don't need to be a maths genius to learn AI. But understanding basic math concepts – especially statistics and probability – helps us understand how AI thinks and takes decisions.

SUB-UNIT 1: IMPORTANCE OF MATH FOR AI

Learning Outcome: Analyze data (numbers/images) and find relation/pattern using Math

1.1 Why is Math Important for AI?

AI works by analyzing data – which can be numbers, images, text, or sound. To analyze this data, AI uses mathematics for:

Finding patterns in large amounts of data.
Making predictions about future events.
Recognizing images and faces.
Understanding language (NLP).
Training AI models to learn from experience.
Measuring accuracy of predictions.
Optimizing – choosing the best solution.
Decision making – weighing options numerically.

When Netflix recommends a movie, it uses math to calculate similarity between your preferences and other users' preferences. When Google Maps shows the fastest route, it uses math to compare thousands of possible paths.

1.2 Uses of Math in AI – The Four Main Branches

Four main branches of mathematics are used in AI:

📊

1. Statistics

Helps in analyzing data, finding averages, trends, and making sense of large data. Used for prediction and decision-making.

🔢

2. Linear Algebra

Deals with vectors, matrices, and equations. Used in image processing, neural networks, and computer vision.

🎲

3. Probability

Helps in predicting the chances of an event happening. Used in risk analysis, weather forecasting, and recommendations.

📈

4. Calculus

Deals with change and motion. Used to train AI models and improve their accuracy (optimization).

🔹 Real-World AI Uses of Each Branch

Branch	Real-World AI Use
Statistics	Spam email detection, sports analytics, customer behaviour analysis.
Linear Algebra	Face recognition, image filters, Google search ranking.
Probability	Weather forecast, Netflix / YouTube recommendations, medical diagnosis.
Calculus	Self-driving cars, deep learning, optimization of AI models.

1.3 Finding Patterns in Numbers and Images

AI is very good at finding patterns in data – something that would take humans a very long time. These patterns help AI make predictions and decisions.

🔢 Number Patterns

A sequence of numbers that follow a specific rule. AI identifies the rule to predict the next number.

Examples: Arithmetic sequences, multiplication tables, Fibonacci series.

🖼️ Picture / Image Patterns

Finding connections, similarities, or rules between images. Used in computer vision to recognize objects and faces.

Examples: Face recognition, object detection, picture analogies.

🔢 Number Patterns (Examples)

Number patterns follow a rule. Look at the sequence and find the next number:

2, 4, 6, 8, 10, ?
(Rule: Add 2. Next = 12)

1, 3, 5, 7, 9, ?
(Rule: Odd numbers. Next = 11)

2, 4, 8, 16, 32, ?
(Rule: Multiply by 2. Next = 64)

1, 1, 2, 3, 5, 8, ?
(Rule: Sum of last two numbers. Next = 13. Fibonacci series!)

🔹 Types of Number Patterns

Arithmetic Pattern: Same number added each time (2, 5, 8, 11, ...).
Geometric Pattern: Same number multiplied each time (3, 9, 27, 81, ...).
Fibonacci Pattern: Next number = sum of last two (1, 1, 2, 3, 5, 8, ...).
Square Numbers: 1, 4, 9, 16, 25, ... (1², 2², 3², ...).
Cube Numbers: 1, 8, 27, 64, 125, ... (1³, 2³, 3³, ...).
Triangular Numbers: 1, 3, 6, 10, 15, ... (sum of natural numbers).

Activity – Observe the Number Pattern: Look at this sequence and find the missing number.
5, 10, 20, 40, 80, ?
Answer: 160 (each number is multiplied by 2).

🖼️ Picture Analogy

Picture Analogy is a type of reasoning where you find the relationship or pattern between a pair of images and apply the same rule to another pair.

🔹 How Picture Analogy Works

The basic format is:

A : B :: C : ?
(A is to B as C is to ?)

Example: 🐄 : 🥛 :: 🐔 : ?
Answer: 🥚 (Cow gives milk, so hen gives egg.)

Example: ☀️ : Day :: 🌙 : ?
Answer: Night

🔹 Uses of Picture Analogy in AI

Teaching AI to identify relationships between objects.
Training computer vision models.
Used in IQ tests and reasoning puzzles.
Helps AI understand similarities between different things.
Foundation of image recognition systems.

Activity – Picture Analogy: Look at pairs of images and find connections between them. Use the pattern to solve similar problems. This is exactly how AI learns visual relationships!

SUB-UNIT 2: STATISTICS

Learning Outcome: Understand the concept of Statistics in real life

2.1 Definition of Statistics

Statistics is the branch of mathematics that deals with the collection, organization, analysis, interpretation, and presentation of data. It helps us understand large amounts of information and make decisions based on it.

🔹 Key Features of Statistics

Deals with numerical data.
Involves collection of facts.
Helps in organizing data systematically.
Used to analyze and interpret data.
Presents data through tables, charts, and graphs.
Helps in decision-making.

2.2 Basic Statistical Measures

In statistics, we commonly use the following measures to describe data:

📊 1. Mean (Average)

Mean = Sum of all values ÷ Number of values

Marks of 5 students: 80, 70, 90, 60, 75
Mean = (80+70+90+60+75) ÷ 5 = 375 ÷ 5 = 75

📊 2. Median (Middle Value)

The middle value when all numbers are arranged in order.

Data: 3, 5, 7, 9, 11
Median = 7 (middle value)

Data (even count): 4, 6, 8, 10
Median = (6+8) ÷ 2 = 7

📊 3. Mode (Most Frequent)

The value that appears most often in data.

Data: 2, 4, 4, 6, 7, 4, 8
Mode = 4 (appears 3 times)

📊 4. Range (Spread of Data)

Range = Highest value − Lowest value

Marks: 45, 78, 60, 92, 55
Range = 92 − 45 = 47

2.3 Applications of Statistics

Statistics is used in almost every field of life. The main applications mentioned in the syllabus are:

🌪️ 1. Disaster Management Statistics helps predict disasters like earthquakes, floods, and cyclones by analyzing past data. Helps plan evacuation, rescue, and relief.

⚽ 2. Sports Used to analyze player performance, team statistics, win/loss ratios, batting averages, goals scored, etc. Used in cricket, football, Olympics.

🏥 3. Disease Prediction Used to predict disease outbreaks, track patients, analyze treatment effectiveness. Very important during COVID-19 pandemic.

☀️ 4. Weather Forecast Uses past weather data to predict future weather – temperature, rain, wind speed. Helps farmers and travellers.

2.4 Other Uses of Statistics in Daily Life

Education: Calculating marks, averages, rankings; analyzing student performance.
Business: Sales analysis, profit / loss calculation, market research.
Government: Census data, population studies, planning policies.
Healthcare: Hospital records, diet planning, medicine effectiveness.
Banking: Interest rates, loans, investments, fraud detection.
Transport: Traffic analysis, accident data, route planning.
Agriculture: Crop yield prediction, soil quality, weather patterns.
Marketing: Customer preferences, ad performance, brand popularity.
Entertainment: TRP ratings, box office collections, music charts.
Social Media: Likes, followers, engagement rates.

Activity 1 – Uses of Statistics in Daily Life: Students collect real-life data (e.g., daily temperature, marks, spending) and apply mean, median, mode, and range to analyze it.

Activity 2 – Car Spotting and Tabulating
Purpose: Implement data collection, analysis, and interpretation.
How to do it:

Stand near a road for 15-30 minutes.
Note down the colour and type (car, bike, bus, truck) of each vehicle passing by.
Prepare a tally table:

Colour Tally Count

Red |||| 4

White |||| || 7

Black ||| 3
Answer questions: Which colour is most common? What is the mean count per minute?
Create a bar graph of your data.

This activity shows how AI collects and analyses data in real life!

Colour	Tally	Count
Red	\|\|\|\|	4
White	\|\|\|\| \|\|	7
Black	\|\|\|	3

SUB-UNIT 3: PROBABILITY

Learning Outcome: Understand the concept of Probability in real life

3.1 Introduction to Probability

Probability is a branch of mathematics that measures the chance or likelihood of an event happening. It tells us how likely something is to occur.

🔹 Key Points About Probability

Probability is always a number between 0 and 1.
0 means the event is impossible (will never happen).
1 means the event is certain (will definitely happen).
Any value in between shows a partial chance.
Can also be expressed as a percentage (0% to 100%).

🔹 Probability Scale

0Impossible

0.25Unlikely

0.5Even Chance

0.75Likely

1Certain

3.2 How to Calculate Probability of an Event

P(Event) = Number of favourable outcomes ÷ Total number of possible outcomes

🔹 Key Terms

Experiment: An activity that has results (e.g., tossing a coin).
Outcome: The result of an experiment (head or tail).
Sample Space: Set of all possible outcomes.
Event: A specific outcome or set of outcomes we are interested in.
Favourable Outcomes: Outcomes that match the event we want.

🔹 Simple Examples

Example 1 – Tossing a coin:
Total outcomes = 2 (Head, Tail)
Favourable outcomes for Head = 1
P(Head) = 1/2 = 0.5 or 50%

Example 2 – Rolling a dice:
Total outcomes = 6 (1, 2, 3, 4, 5, 6)
Favourable outcomes for getting "4" = 1
P(4) = 1/6 ≈ 0.167 or 16.67%

Example 3 – Rolling a dice for an even number:
Total outcomes = 6
Even outcomes = 3 (2, 4, 6)
P(even) = 3/6 = 1/2 = 0.5 or 50%

Example 4 – Picking a red ball from a bag with 3 red and 5 blue balls:
Total balls = 8
Favourable (red) = 3
P(red) = 3/8 = 0.375 or 37.5%

3.3 Types of Events in Probability

1. Sure / Certain EventAn event that will definitely happen. Probability = 1.
Example: The sun rising tomorrow.

2. Impossible EventAn event that cannot happen. Probability = 0.
Example: Getting 7 on a normal dice.

3. Simple EventAn event with only one outcome.
Example: Getting "Head" on a coin toss.

4. Compound EventAn event with more than one outcome.
Example: Getting an even number (2, 4, 6) on a dice.

5. Independent EventsEvents that do NOT affect each other.
Example: Tossing a coin twice.

6. Dependent EventsEvents that affect each other.
Example: Drawing 2 cards without replacement.

7. Mutually Exclusive EventsEvents that cannot happen together.
Example: Getting Head AND Tail in one toss.

8. Equally Likely EventsEvents with same probability.
Example: Getting Head or Tail on a fair coin.

9. Complementary EventsOne event happens OR the other; P(A) + P(not A) = 1.
Example: Pass or Fail an exam.

10. Exhaustive EventsAll possible outcomes combined.
Example: All 6 faces of a dice.

Exercise – Identify the Type of Event:

Getting a number less than 7 on a dice → Sure Event
A dog laying eggs → Impossible Event
Getting a "3" on a dice → Simple Event
Getting an odd number on a dice → Compound Event
Winning a cricket match and losing it → Mutually Exclusive

3.4 Applications of Probability in Real Life

Probability is used in many real-life and AI applications:

⚽ 1. Sports Predicting team winning chances, player performance, match outcomes. IPL uses probability to predict team wins.

☁️ 2. Weather Forecast "30% chance of rain today" – this is probability! Based on past weather data.

🚦 3. Traffic Estimation Google Maps calculates the probability of reaching destination at a given time based on traffic.

🏥 4. Medical Diagnosis Doctors use probability to estimate chances of disease based on symptoms.

🔹 Other Real-Life Uses

Insurance: Calculating premiums based on risk.
Gambling and Games: Cards, lottery, dice games.
Stock Market: Predicting share prices.
Business: Risk analysis, product launch decisions.
Space Research: Calculating mission success rates.
Quality Control: Chance of a product being defective.
AI / Machine Learning: Predicting user behaviour, classification.
Online Recommendations: Probability that you'll like a product/movie.

AI + Probability: Most AI models give answers in terms of probability. For example, a spam filter says "95% chance this is spam" – that's probability working inside AI!

3.5 Practice Problems (Revision Time)

Q1: A bag contains 4 red balls and 6 green balls. What is the probability of picking a green ball?
Solution: P(green) = 6/10 = 3/5 = 0.6 or 60%

Q2: When two coins are tossed together, what is the probability of getting two heads?
Solution: Total outcomes = HH, HT, TH, TT = 4
Favourable (HH) = 1
P(two heads) = 1/4 = 0.25 or 25%

Q3: A card is drawn from a deck of 52 cards. What is the probability of drawing an ace?
Solution: Total cards = 52
Aces = 4
P(ace) = 4/52 = 1/13 ≈ 0.077 or 7.7%

Q4: In a box, there are 10 mangoes of which 3 are rotten. Find the probability of picking a good mango.
Solution: Total = 10, Good = 10 − 3 = 7
P(good) = 7/10 = 0.7 or 70%

Quick Revision – Key Points to Remember

Math is the language of AI – it helps AI analyze, predict, and decide.
4 Branches used in AI: Statistics, Linear Algebra, Probability, Calculus.
Number Patterns: Arithmetic, Geometric, Fibonacci, Square, Cube, Triangular.
Picture Analogy: A : B :: C : ? (finding relationship between image pairs).
Statistics = collection, organization, analysis, interpretation, and presentation of data.
Statistical Measures: Mean (average), Median (middle), Mode (most frequent), Range (spread).
Mean: Sum of values ÷ Number of values.
Range: Highest value − Lowest value.
Applications of Statistics: Disaster Management, Sports, Disease Prediction, Weather Forecast.
Probability = measure of how likely an event is to happen (0 to 1).
P(Event) = Favourable outcomes ÷ Total outcomes.
Probability = 0 → Impossible; Probability = 1 → Certain.
Types of Events: Sure, Impossible, Simple, Compound, Independent, Dependent, Mutually Exclusive, Equally Likely, Complementary, Exhaustive.
Applications of Probability: Sports, Weather, Traffic, Medical Diagnosis, Insurance, AI predictions.
AI uses Probability to give predictions (e.g., 95% spam, 80% cat).

🧠Practice Quiz — test yourself on this chapter→