MATH FOR AI R.SUCHITRA, MCA, M.ED.

,
(STATISTICS & DAV PUBLIC SCHOOL,
VELACHERY, CHENNAI

PROBABILITY)
HOW ARE MATH AND AI
RELATED?
Math is the study of patterns
To solve the puzzles, you identify an
order/arrangement in the list of numbers or
the images.
 This arrangement is called a pattern.
These patterns exist all around us.
We have patterns in numbers, images, and
language.
 KEY MATHEMATICAL
AREAS
Linear Algebra: IN AI
Used in processing images, creating neural
networks, and handling data in large matrices.
Probability and Statistics: Essential for making predictions,
dealing with uncertainties, and handling data patterns.
Calculus: Plays a key role in optimizing algorithms, especially
in training machine learning models through gradient
descent.
Discrete Mathematics: Important for algorithms, logic, and
understanding complex data structures.
 AI USES MATH FOR:
Statistics (Exploring data): Example – What is the middle
value of the data? Which is the most common value in the
data?

Calculus (training and improving AI model): Example –

which line is more slanted? Which figure covers more area?

Linear Algebra (finding out unknown or missing values):

Example – How many plants are there in total? How many
cars are there in a city?

Probability (predicting different events): Example – what

will be the possible results of a coin toss? Will it rain
AI IS A WAY TO
RECOGNIZE
 PATTERNS
AI can learn to recognize
patterns, like human beings.

AI can see patterns in different

types of data - numbers, images,
and speech and text.

These patterns help AI to solve

puzzles – like identifying dogs
and muffins, or predicting
CAN YOU IDENTIFY ANY
PATTERN IN THE
Math IMAGE
is the study of
GIVEN BELOW?
patterns

AI is a way to recognize

patterns in order to take
decisions

AI needs Math to study and

recognize patterns in order
to take decisions
STATISTICS
Statistics is used for collecting, exploring,
and analyzing the data. It also helps in
Data is collected from various sources.
drawing conclusions from data.
Data is explored and cleaned to be used.

Analysis of data is done to understand it better.

Conclusions and decisions can be made from the

data.
STATISTICAL METHODS IN
AI
Regression Analysis:
Predicts relationships between variables.
Applications: Trend prediction, recommendation
systems.
Bayesian Statistics:
Updates probabilities with new evidence.
Applications: Spam detection, medical diagnoses.
Machine Learning Algorithms:
Uses statistical techniques for pattern recognition and
predictions.
Examples: Decision Trees, SVMs, KNN.
APPLICATIONS OF
STATISTICS:
Predict the performance ofsports teams

 It can be used to find out specific things

such as:
 the reading level of students
 the opinions of voters
 the average weight of a city’s resident
GENERAL APPLICATIONS
OF STATISTICS
 DISASTER
MANAGEMENT:
Risk Assessment:
Analyzes historical data to predict
potential disasters.
Identifies high-risk areas and
populations.
Resource Allocation:
Optimizes the distribution of
emergency supplies.
Ensures timely deployment of
rescue teams.
Damage Assessment:
Uses AI to estimate damage
through satellite images and drone
data.
Speeds up recovery planning and
relief efforts.
 DISEASE PREDICTION
Epidemiological Studies:
Analyzes disease patterns and
risk factors.
Helps in understanding spread
and control strategies.
Surveillance Systems:
Monitors health data in real time.
Detects early signs of outbreaks.
Predictive Modeling:
Uses AI to forecast disease
spread based on data trends.
Supports targeted prevention
efforts.
Forecasting Outbreaks:
Predicts future outbreaks using
weather, population, and mobility
data.
WEATHER FORECASTING
Data Analysis:
 Processes data from
satellites, sensors, and
weather stations.
 Identifies patterns in
temperature, humidity, and
wind behavior.
Forecast Models:
 Uses AI models to predict
weather trends and extreme
conditions.
 Enhances accuracy for
 CENTRAL TENDENCY
Mean:
 The average of all data points.
 Formula: Sum of values ÷ Number
of values.
Median:
 The middle value in a sorted
dataset.
 Divides data into two equal
halves.
Mode:
 The value that appears most
frequently in the dataset.
MEAN
Definition:
 The mean is the average of a set of numbers.
 It represents a central value of the dataset.

Formula:
 Mean=Sum of all values / Total number of values
Example:
 Dataset: 5, 8, 12, 10, 15
 Mean=(5+8+12+10+15)/5=50/5=10
 MEAN FROM GROUPED
DATA
Formula for Mean from Steps:
Grouped Data:  Calculate the midpoint (x) of
Mean = ∑(f×x)/N each class interval.
Where:  Multiply each midpoint by
 f = frequency of the class the corresponding frequency
(f).
 x = midpoint of the class  Sum up the products of f×x.
interval
 Divide the total sum by the
 N = total frequency (sum
total frequency (N).
of all frequencies)
EXAMPLE FOR MEAN FROM GROUPED DATA
Class Frequency Midpoint
Interval (f) (x) f×x
0-10 5 5 25
10-20 8 15 120
20-30 12 25 300
30-40 10 35 350
Given grouped data:
F × x = 25 + 120 + 300 + 350 = 795
Total frequency
N = 5 + 8 + 12 + 10 = 35
Mean = 795/35 = 22.71
MEDIAN
Definition:
The median is the middle value in a sorted dataset,
which divides the data into two equal halves.
Steps to Find the Median:
Sort the data in ascending or descending order.
If the number of data points (n) is odd:
The median is the middle value, i.e., the value at
position (n+1)/2​.
If the number of data points (n) is even:
The median is the average of the two middle values,
i.e., the average of the values at positions (n/2)​
and (n+1)/2
EXAMPLE FOR MEDIAN
Odd number of data points:
Dataset: 3, 5, 8, 12, 15
Median = 8 (middle value)
Even number of data points:
Dataset: 3, 5, 8, 12
Median = (5+8)/2=6.5
 MEDIAN FROM GROUPED
DATA
Formula for Median from Steps to Calculate Median:
Grouped Data:
Median=L+ x h  Determine the median class:

where: Find the class interval where

 L = Lower boundary of the
the cumulative frequency
median class
 N = Total frequency (sum of all exceeds N/2
frequencies)
F = Cumulative frequency  Apply the formula: Use the
before the median class values from the median class
 f = Frequency of the median
class to calculate the median.
 h = Class width (difference
between upper and lower
EXAMPLE FOR MEDIAN FROM GROUPED DATA
Cumulative
Class Interval Frequency (f)
Frequency (CF)
0-10 5 5
10-20 8 13
20-30 12 25
30-40 10 35

•Total frequency N=35

•Median class: 20-30 (because N/2=17.5/2, and CF of 13 <
17.5 < 25)
•Lower boundary L=20, Frequency f=12, Cumulative
frequency before the median class F=13, Class width h=10
Applying the formula:
Median=20+(×10=20+()×10
= 20+3.75 = 23.75
MODE
The mode is the value or class interval that appears most frequently in
a dataset.
Inspection Method
Identify the value that appears most frequently in the dataset.
Example: Dataset: 2, 4, 4, 6, 8 → Mode = 4 (appears twice).
Mode for Frequency Distribution
Formula: Mode = L+ x h
L: Lower boundary of the modal class
𝑓1​: Frequency of the modal class
𝑓0: Frequency of the class before the modal class
𝑓2: Frequency of the class after the modal class
ℎ: Class width
STEPS FOR FREQUENCY DISTRIBUTION
Identify the modal class (class with the highest frequency).
Apply the formula to find the mode.
Example: Class Interval Frequency (f)
0-10 5
10-20 8
20-30 12
30-40 7

L=20, 𝑓1=12, 𝑓0=8, 𝑓2=7, ℎ=10

Mode = 20+×10 = 20+(4/9)×10 = 24.44
PROBABILITY
TO UNDERSTAND THE POSSIBILITY
OF OCCURRENCE OF AN EVENT
INTRODUCTION TO
PROBABILITY
 Probability is a way to tell us how likely something is to
happen.
 For example – When a coin is tossed, there are two
possible results or outcomes: heads (H) or tails (T)
 The probability equation defines the likelihood of the
happening of an event.
 It is the ratio of favorable outcomes to the total favorable
outcomes.
 The probability formula can be expressed as,
Probability of an Event =
WAYS TO EXPRESS
PROBABILITY
 Certain events: An event will happen
without a doubt
 Likely events: The probability of one event
is higher than the probability of another
event
 Unlikely events: One event is less likely to
happen than another event
 Impossible events: There's no chance of an
event happening
 Equal Probability events: Chances of each
 The probability of an event occurring is
somewhere between impossible and certain.
 If an event is certain or sure to happen, it will
have a probability of 1.
 For example, the probability that it will rain in
the state of Florida at least once in a specific
year is 1.
 If an event will never happen or is impossible,
it will have a probability of 0.
 For example, the probability that you can pick a
red ball from a bag containing only blue balls is
0.
PROBABILITY -
APPLICATIONS
SPORTS
Probability can be used in
estimating batting average in
Cricket.
Batting average in Cricket
represents how many runs a
batsman would score before getting
out.
For instance, if a batsman had
scored 45 runs out of 100 from only
boundaries in the last match. Then,
there is a chance that he will score
45% of his runs in the next match
from boundaries.
WEATHER FORECASTING

One of the most common real-life examples of using

probability is weather forecasting.
It is used by weather forecasters to assess how likely it i
that there will be rain, snow, clouds, etc., on a given da in
a certain area.
Forecasters may say things like “there is a 70% chance of
rain today between 4 PM and 6 PM” to indicate a medium
to high likelihood of rain during certain hours.
TRAFFIC ESTIMATION
Regular people often use probability
when they decide to drive to
someplace.
Based on the time of day, location in
the city, weather conditions, etc.
people tend to make probability
predictions about how bad traffic will
be during a certain time.
For example, if you think there’s a
90% probability that traffic will be
heavy from 6 PM to 7:30 PM in your
vicinity then you may decide to wait
during that time.
THANK YOU