10-09-2025

Supervised Learning

[Link] M.E.,Ph.D
[ Top 2% Scientist in the World ]
Associate Professor
Department of Computer Science and Engineering Education,
National Institute of Technical Teachers Training and Research (NITTTR), Bhopal.
E-Mail: sganapathy@[Link], [Link]@[Link]
Phone: +91 94888 69712
“Hard Work Never Fail” & “Continuous Effort will bring you a Great Height”

9/10/2025 [Link], Associate Professor, DCSEE, NITTTR, Bhopal, MP, INDIA-462002. 1

OUTLINE
• AI Vs ML Vs DL
• How does AI working?
• AI & ML
• What is Machine Learning?
• How does ML is working?
• Supervised Learning Algorithms
• Linear Regression
• Naïve Bayes
• Random Forest
• Decision Tree
• MLP
• SVM
• K-NN classifier
• ANN
9/10/2025 [Link], Associate Professor, DCSEE, NITTTR, Bhopal, Madhya Pradesh, INDIA-462002. 2

1
 10-09-2025

Outcomes
Learners are able to explain the various supervised learning
algorithms and also to apply them in various applications.

9/10/2025 [Link], Associate Professor, DCSEE, NITTTR, Bhopal, Madhya Pradesh, INDIA-462002. 3

AI Vs ML Vs DL
Artificial Intelligence (AI)
Enable a machine to think as human

Machine Learning (ML)

Supervised, Unsupervised & Semi-Supervised (Reinforcement)
Explore and Analyze the data through Statistical tools

Deep Learning (DL)

Multi Neural Network Architecture
ANN, CNN, RNN

2
 10-09-2025

How does AI is working?

• Thinking humanly – think as a human being

• Thinking rationally – think logical manner

• Acting humanly - Act as a human being

• Acting rationally – Act with logic

AI & ML

• "Artificial intelligence is a set of algorithms and intelligence to try to

mimic human intelligence. Machine learning is one of them, and deep
learning is one of those machine learning techniques."

3
 10-09-2025

What is Machine Learning?

• The capability of Artificial Intelligence systems to learn by extracting patterns

from data is known as Machine Learning.

• Machine Learning is an idea to learn from examples and experience, without

being explicitly programmed. Instead of writing code, you feed data to the
generic algorithm, and it builds logic based on the data given.

• The main purpose of machine learning is to explore and construct algorithms

that can learn from the previous data and make predictions on new input data.

How does ML is working?

• Supervised Learning - Handle the labeled data (training data with result
column)

• Unsupervised Learning – Handle the unlabeled data (test data without result
column)

• Semi-supervised Learning (Reinforcement Learning)

- Handle the Labeled and Unlabeled data

- 50% data with result column & 50% data without result column
- According to the training experience with first part of data

4
 10-09-2025

How does Supervised ML is working?

Supervised Learning Algorithms

• Classification: The machine learning program must draw a conclusion from
observed values. For example, when filtering emails as ‘spam’ or ‘not spam’, the
program must look at existing observational data and filter the emails
accordingly.
• Regression: The machine learning program must estimate the relationships
among variables. Regression analysis focuses on one dependent variable and a
series of other changing variables - making it particularly useful for prediction
and forecasting.
• Forecasting: The process of making predictions about the future based on the past
and present data, and is commonly used to analyse trends.
10

5
 10-09-2025

Supervised Learning Algorithms

1. Linear Regression
2. Naïve Bayes
3. Random Forest
4. Decision Tree
5. Multilayer Perceptron (MLP)
6. Support Vector Machine (SVM)
7. K-Nearest Neighbor classifier (k-NN)
8. Artificial Neural Network (ANN)

11

Linear Regression

12

6
 10-09-2025

Regression Analysis
Need for Regression Analysis in Machine Learning
 Regression Analysis is used to find the relationship between the
dependent variable and independent variable that are real or continuous
data.
 Here, more than two variables are also to be considered for analysis.
 The regression is used to predict the future by analyzing the past records
and the data trends.
 The analysis is to be done based on the relationship between the
dependent and independent variables of the past records.
 This regression is useful for the data analyst and scientist to predict the
future effectively by referring the regression analysis.
13

Linear Regression
 Linear Regression finds the relationship
between the dependent and independent
variables of the given input data.

 There are two types of Linear Regression.

[Link]/ Simple Linear Regression: It

considers only one independent variable.
[Link] / Multiple Linear Regression:
It considers minimum of two
independent variables.

14

7
 10-09-2025

Linear Regression Line

• It represents the relationship difference between the
dependent variables and independent variables.
• Dependent variable represents the predictable (target)
variable.
• Independent variable means that the variable
(predictor) used to do the prediction.
• The relationship differences can be categorized into
two such as positive linear and negative linear.
• Positive Linear: Increases the number of dependent
variables in Y-axis and also increases the number of
independent variables in X-axis.
• Negative Linear: Decreases the number of dependent
variables in Y-axis and also increases the number of
independent variables in X-axis.
15

Linear Regression
Advantages:
 It consumes less computational time to resolve complex problems.
 It consumes less time to train any large dataset.
 It works based on the assumptions and also proved the prediction result
through statistical method with ease of understanding.
 It is useful for predicting the future of various environments including biology,
social science, business, etc.

Disadvantages:

 Relationships between the variables are not capable of doing perfect prediction.
Because, the average relationship between the variables may not be the
description of them.
16

8
 10-09-2025

Differentiate Simple and Multiple Linear

Regression
Simple Linear Regression Multiple Linear Regression

Uses one dependent variable and one independent Uses one dependent variable and minimum two
variable to find the relationship independent variables to find the relationship
Complexity is low due to the consideration of one Complexity is high due to the consideration of more
relationship relationships
Visualize with 2D 3D is required for visualization

Low Risk due to the presence of single predictor Risk is high due to the presence of more predictors

Easy to understand the model Difficult to understand than simple linear regression

Simple predictions can be done Complex and Multifactorial predictions can be done

17

Linear Regression – An Example

 Let us consider an example where the five weeks sales data (in thousands) is
given as shown in table 1. Apply linear regression technique to predict the 6th
and 8th week sales.
 Sample data is shown in Table 1.

𝒙𝒊 (Week) 𝒚𝒊 (Sales in thousands) SALES DETAILS

8.9
8.1
Sales (in thousands)

1 4.3
5.7
2 4.9 4.3
4.9

3 5.7
4 8.1
1 2 3 4 5
5 8.9 Week
18

9
 10-09-2025

Linear Regression – An Example

 Solution: Here, there are 5 items, i.e., i = 1, 2, 3, 4 and 5. The computation table
is shown in table 2.
 Here, there are 5 samples, so i ranges from 1 to 5.

𝒙𝒊 𝒚𝒊 (𝒙𝒊 )𝟐 𝒙𝒊 × 𝒚𝒊
1 4.3 1 4.3

2 4.9 4 9.8

3 5.7 9 17.1

4 8.1 16 32.4

5 8.9 25 44.5
Sum= 15 Sum = 31.9 Sum= 55 Sum = 108.1
Average of 𝒙𝒊 Average of 𝒚𝒊 Average of 𝒙𝒊 𝟐 Average of 𝒙𝒊 × 𝒚𝒊
𝒙= 3 𝒚= 6.38 𝒙𝒊 𝟐 = 11 𝒙𝒊 × 𝒚𝒊 = 21.62 19

Linear Regression – An Example

 Let us compute the slope and intercept now using the below equations:

( ̅ )( )
𝒂𝟏 =
( ) ( ̅)
. .
𝒂𝟏 = = 1.24
𝒂𝟎 = (𝑦)– 𝑎 × (𝑥̅ )

𝒂𝟎 = 6.38 – 1.24 × 3 = 2.66

 Now, the regression line is by considering the these values.

𝑦 = 2.66x + 1.24 20

10
 10-09-2025

Linear Regression – An Example

 Let us model the relationship as 𝑦 = 𝒂𝟎 + 𝒂𝟏 × x
 Therefore, the fitted line for the above data is: y = 2.66 + 1.24x
 The predicted 6th week sale would be (when x=6),
 y = 2.66 + 1.24 × 6
 = 10.1
 The predicted 8th month would be (when x=8)
 y = 2.66 + 1.24 × 8
 = 12.58
 All sales are in thousands.

21

Linear Regression – An Example

 After predicting the sales of 6th week and 8th week by considering the past
five weeks sales, we will get the following graph.
SALES PREDICTION USING LINEAR REGRESSION

12.58
Sales (in thousands)

10.1
8.9
8.1
5.7
4.3 4.9

1 2 3 4 5 6 8
Week
22

11
 10-09-2025

Naïve Bayes Classifier

23

Naïve Bayes Classifier

What is Naïve Bayes Classifier?
 Naïve Bayes (NB) algorithm is a supervised learning algorithm,
which is based on Bayes Theorem and used for solving classification
problems.
 NB is mainly used for performing text classification.
 It is a simple and effective classifier
 It helps to build the fast machine learning models for performing the
fast predictions.
 NB is a probabilistic classifier and it is used to perform prediction
process based on an object’s probability.
 Examples: Spam filtration, Sentimental Analysis, and Article
classification.
24

12
 10-09-2025

Bayes Theorem
• Bayes’ theorem is also known as Bayes’ Rule or Baye’s law, which is used to
determine the probability of a hypothesis with prior knowledge. It depends on
the conditional probability.
• The formula for Bayes’ theorem is given as:
𝑃 𝐵 𝐴 . 𝑃(𝐴)
𝑃 𝐴𝐵 =
𝑃 𝐵 . 𝑃(𝐴|𝐵)

Where,
• P(A|B) is a POSTERIOR (The probability of “A” being True, given “B” is True.)
• P(B|A) is a LIKELIHOOD (The probability of “B” being True, given “A” is True.)
• P(A) is a PRIOR (The probability “A” being True. This is the knowledge.)
• P(B) is a MARGINALIZATION (The probability “B” being True.)
25

Naïve Bayes Classifier – An Example

• Problem: If the weather is sunny, then the Player should play or not?
• Solution: To solve this, first consider the below dataset.

Weather Yes No

Overcast 5 0
Rainy 2 2
Sunny 3 2
Total 10 4

26

13
 10-09-2025

Naïve Bayes Classifier – An Example

• Find the probability of Overcast, Rainy and Sunny, also the
probability of being Yes and No.

Weather Yes No Probability

Overcast 5 0 5 + 0 / 14 = 5 / 14 = 0.35
Rainy 2 2 2 + 2 / 14 = 4 / 14 = 0.29
Sunny 3 2 3 + 2 /14 = 5 / 14 = 0.35
All 10/14 = 0.71 4 / 14 = 0.29
27

Naïve Bayes Classifier – An Example

Apply Bayes Theorem
• P(Yes|Sunny) = P(Sunny|Yes)*P(Yes)/P(Sunny)
• P(Sunny|Yes) = 3/10 = 0.3
• P(Sunny) = 0.35
• P(Yes) = 0.71
• P(No|Sunny) = P(Sunny|No)*P(No)/P(Sunny)
• P(Sunny|No) = 2/4 = 0.5
• P(Sunny) = 0.35
• P(NO|Sunny) = (0.5*0.29)/0.35 = 0.41
• So as we can see from the above calculation that
P(Yes|Sunny)>P(No|Sunny)
• Hence on a Sunny day, Player can play the game.
28

14
 10-09-2025

Advantages & Disadvantages of Naïve Bayes

Classifier
Advantages:
 Naïve Bayes’ is one of the fast and easy ML algorithms to predict a
class of datasets.
 It can be used for binary as well as Multi-class classifications.
 It performs well in Multi-class predictions as compared to the other
algorithms.
 It is the most popular choice for text classification problems.

Disadvantages:
 Naïve Bayes assumes that all features are independent or unrelated,
so it cannot learn the relationship between features.
29

Applications of Naïve Bayes Classifier

 It is used for credit scoring.

 It is used in medical data classification.

 It can be used in real-time predictions because Naïve Bayes classifier

is an eager learner.

 It is used in Text classification such as Spam filtering and Sentiment

Analysis.

30

15
 10-09-2025

Random Forest

31

Random Forest Classifier

 Random Forest is a famous machine learning algorithm that uses
supervised learning methods.
 You can apply it to both classification and regression problems.
 It is based on ensemble learning, which integrates multiple classifiers to
solve a complex issue and increases the model's performance.
 In layman's terms, Random Forest is a classifier that contains several
decision trees on various subsets of a given dataset and takes the average
to enhance the predicted accuracy of that dataset.
 Instead of relying on a single decision tree, the random forest collects
the result from each tree and expects the final output based on the
majority votes of predictions. 32

16
 10-09-2025

Working flow of Random Forest Classifier

Step 1: Pick M data points at random from the training set.

Step 2: Create decision trees for your chosen data points (Subsets).

Step 3: Each decision tree will produce a result. Analyze it.

Step 4: For classification and regression, accordingly, the final output

is based on Majority Voting or Averaging, accordingly.

33

Random Forest Classifier – An Example

Consider the following scenario: A dataset containing several fruits

images. And the Random Forest Classifier is given this dataset. Each
decision tree is given a subset of the dataset to work with. During the
training phase, each decision tree generates a prediction result. The
Random Forest classifier predicts the final decision based on most
outcomes when a new data point appears.

34

17
 10-09-2025

Random Forest Classifier – An Example

35

Essential of Random Forest Classifier

Diversity- When creating an individual tree, not all qualities, variables, or features
are taken into account; each tree is unique.

Immune to dimensionality constraint- The feature space is minimized because

each tree does not consider all features.

Parallelization- Each tree is built from scratch using different data and properties.
This means we can fully utilize the CPU to create random forests.

Train-Test split- In a random forest, there is no need to separate the data for train
and test because the decision tree will always miss 30% of the data.

Stability- The result is stable because it is based on majority voting/averaging.

36

18
 10-09-2025

Applications of Random Forest Classifier

Banking: It is mainly used in the banking industry to identify loan risk.

Medicine: To identify illness trends and risks.

Land Use: Random Forest Classifier is also used to classify places with
similar land-use patterns.

Market Trends: You can determine market trends using this algorithm.

37

Decision Tree

38

19
 10-09-2025

Introduction to Decision Tree

 Decision tree is a simple diagram that shows different choices and

their possible results helping you make decisions easily.

 A decision tree is a graphical representation of different options for

solving a problem and show how different factors are related.

 It has a hierarchical tree structure starts with one main question at the
top called a node which further branches out into different possible
outcomes where:
39

Introduction to Decision Tree

 Root Node is the starting point that represents the entire dataset.

 Branches: These are the lines that connect nodes. It shows the flow
from one decision to another.

 Internal Nodes are Points where decisions are made based on the
input features.

 Leaf Nodes: These are the terminal nodes at the end of branches that
represent final outcomes or predictions
40

20
 10-09-2025

Workflow of Decision Tree

41

Decision Tree – An Example

 Problem: Predicting Whether a Person Likes Computer Games.
Imagine you want to predict if a person enjoys computer games based
on their age and gender.

 Solution:

42

21
 10-09-2025

Decision Tree – An Example

• Start with the Root Question (Age):
• The first question is: “Is the person’s age less than 15?”
• If Yes, move to the left.
• If No, move to the right.
• Branch Based on Age:
• If the person is younger than 15, they are likely to enjoy computer games (+2
prediction score).
• If the person is 15 or older, ask the next question: “Is the person male?”
• Branch Based on Gender (For Age 15+):
• If the person is male, they are somewhat likely to enjoy computer games (+0.1
prediction score).
• If the person is not male, they are less likely to enjoy computer games (-1
prediction score)
43

Decision Tree – An Example

44

22
 10-09-2025

Decision Tree – An Example

• Example: Predicting Whether a Person Likes Computer Games Using Two Decision Trees

• Tree 1: Age and Gender

• The first tree asks two questions:

• “Is the person’s age less than 15?”

• If Yes, they get a score of +2.

• If No, proceed to the next question.

• “Is the person male?”

• If Yes, they get a score of +0.1.

• If No, they get a score of -1.

45

Decision Tree – An Example

• Tree 2: Computer Usage
• The second tree focuses on daily computer usage:
• “Does the person use a computer daily?”
• If Yes, they get a score of +0.9.
• If No, they get a score of -0.9.
• The final prediction score is the sum of scores from both trees
• Information Gain and Gini Index in Decision Tree
• Till now, we have discovered the basic institution and approach of how decision
tree works, so lets just move to the attribute selection measure of decision tree.
• We have two popular attribute selection measures used:
• 1. Information Gain 2. Gini Index

46

23
 10-09-2025

Decision Tree – An Example

• Information Gain:

• Where, S is a set of instances, A is an attribute, Sv is the subset of S with A = v, and

Values (A) is the set of all possible values of A, then
• Example:
• For the set X = {a,a,a,b,b,b,b,b}
• Total instances: 8
• Instances of b: 5
• Instances of a: 3

47

Advantages of Decision Tree

 Simplicity and Interpretability: Decision trees are straightforward and easy to
understand. You can visualize them like a flowchart which makes it simple to see
how decisions are made.

 Versatility: It means they can be used for different types of tasks can work well
for both classification and regression

 No Need for Feature Scaling: They don’t require you to normalize or scale your
data.

 Handles Non-linear Relationships: It is capable of capturing non-linear

relationships between features and target variables. 48

24
 10-09-2025

Disadvantages of Decision Tree

 Overfitting: Overfitting occurs when a decision tree captures noise and details in
the training data and it perform poorly on new data.

 Instability: instability means that the model can be unreliable slight variations in
input can lead to significant differences in predictions.

 Bias towards Features with More Levels: Decision trees can become biased
towards features with many categories focusing too much on them during
decision-making.

 This can cause the model to miss out other important features led to less accurate
predictions.
49

Applications of Decision Tree

 Loan Approval in Banking:

 Medical Diagnosis

 Predicting Exam Results in Education

 Attack Prediction

50

25
 10-09-2025

References
• [Link]
• [Link]
• [Link]

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 51

THANK YOU

9/10/2025 [Link], Associate Professor, DCSEE, NITTTR, Bhopal, Madhya Pradesh, INDIA-462002. 52

26
 10-09-2025

Unsupervised Learning

[Link] M.E.,Ph.D
[ Top 2% Scientist in the World ]
Associate Professor
Department of Computer Science and Engineering Education,
National Institute of Technical Teachers Training and Research (NITTTR), Bhopal.
E-Mail: sganapathy@[Link], [Link]@[Link]
Phone: +91 94888 69712
“Hard Work Never Fail” & “Continuous Effort will bring you a Great Height”

9/10/2025 [Link], Associate Professor, DCSEE, NITTTR, Bhopal, MP, INDIA-462002. 53

SESSION-2
OUTCOME: Learners are able to explain the various clustering algorithms and also to
apply them in various applications

9/10/2025 [Link], Associate Professor, DCSEE, NITTTR, Bhopal, Madhya Pradesh, INDIA-462002. 54

27
 10-09-2025

AI Vs ML Vs DL
Artificial Intelligence (AI)
Enable a machine to think as human

Machine Learning (ML)

Supervised, Unsupervised & Semi-Supervised (Reinforcement)
Explore and Analyze the data through Statistical tools

Deep Learning (DL)

Multi Neural Network Architecture
ANN, CNN, RNN

55

How does AI is working?

• Thinking humanly – think as a human being

• Thinking rationally – think logical manner

• Acting humanly - Act as a human being

• Acting rationally – Act with logic

56

28
 10-09-2025

AI & ML

• "Artificial intelligence is a set of algorithms and intelligence to try to

mimic human intelligence. Machine learning is one of them, and deep
learning is one of those machine learning techniques."

57

What is Machine Learning?

• The capability of Artificial Intelligence systems to learn by extracting patterns

from data is known as Machine Learning.

• Machine Learning is an idea to learn from examples and experience, without

being explicitly programmed. Instead of writing code, you feed data to the
generic algorithm, and it builds logic based on the data given.

• The main purpose of machine learning is to explore and construct algorithms

that can learn from the previous data and make predictions on new input data.

58

29
 10-09-2025

How does ML is working?

• Supervised Learning - Handle the labeled data (training data with result
column)

• Unsupervised Learning – Handle the unlabeled data (test data without result
column)

• Semi-supervised Learning (Reinforcement Learning)

- Handle the Labeled and Unlabeled data

- 50% data with result column & 50% data without result column
- According to the training experience with first part of data

59

Introduction to Unsupervised Learning

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 60

30
 10-09-2025

Introduction to Unsupervised Learning

 Unsupervised learning is where you only have input data (X) and no corresponding output variables.
 The goal for unsupervised learning is to model the underlying structure or distribution in the data
in order to learn more about the data.
 These are called unsupervised learning because unlike supervised learning above there is no correct
answers and there is no teacher. Algorithms are left to their own devises to discover and present the
interesting structure in the data.
 Supervised: All data is labeled and the algorithms learn to predict the output from the input data.
 Unsupervised: All data is unlabeled and the algorithms learn to inherent structure from the input
data.
 Semi-supervised: Some data is labeled but most of it is unlabeled and a mixture of supervised and
unsupervised techniques can be used.

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 61

Introduction to Unsupervised Learning

 Unsupervised Learning
 Training samples contain only input patterns
 No desired output is given (teacher-less)
 Learn to form classes/clusters of sample patterns according to similarities among them
 Patterns in a cluster would have similar features
 No prior knowledge as what features are important for classification, and how many
classes are there.
 It is often easier to obtain unlabeled data from a lab / instrument or a computer than labeled data,
which can require human intervention.
 For example it is difficult to automatically assess the overall sentiment of a movie review: is it favorable
or not?
9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 62

31
 10-09-2025

Introduction to Unsupervised Learning

Distance Measurement Formulae:
1. Euclidean distance
2. Manhattan distance
3. Mahalanobis distance
4. Minkowski distance
5. Chebychev distance

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 63

Clustering Techniques
 Types:
1. Exclusive Clustering: K-means
2. Overlapping Clustering: Fuzzy C-means
3. Hierarchical Clustering: Agglomerative clustering, divisive clustering
4. Probabilistic Clustering: Mixture of Gaussian models

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 64

32
 10-09-2025

Clustering Techniques
 Exclusive Clustering: K-means : randomly initialize the k cluster centers, and
iterate between the two steps we just saw.
1. Randomly initialize the cluster centers, c1, ..., cK
2. Given cluster centers, determine points in each cluster
For each point p, find the closest ci. Put p into cluster i
3. Given points in each cluster, solve for ci
Set ci to be the mean of points in cluster i
4. If ci have changed, repeat Step 2

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 65

Clustering Techniques
Exclusive Clustering
 Pros
 Simple, fast to compute
 Converges to local minimum of within-cluster squared error
 Cons
 Setting k?
 Sensitive to initial centers
 Sensitive to outliers
 Detects spherical clusters
 Assuming means can be computed

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 66

33
 10-09-2025

Clustering Techniques
 Overlapping Clustering: Fuzzy C-means : One data point may belong to two or
more cluster with different memberships.
 Objective function:

 where 1 ≤ m< ∞
 An extension of k-means

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 67

Clustering Techniques
Overlapping Clustering:
Pros:
 Allows a data point to be in multiple clusters
 A more natural representation of the behavior of genes
- genes usually are involved in multiple functions
 Cons:
 Need to define c (k in K-means), the number of clusters
 Need to determine membership cutoff value
 Clusters are sensitive to initial assignment of centroids
- Fuzzy c-means is not a deterministic algorithm

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 68

34
 10-09-2025

Clustering Techniques
Hierarchical Clustering:

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 69

Clustering Techniques

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 70

35
 10-09-2025

Clustering Techniques
Agglomerative Clustering:
 It is more popular then divisive methods.
 At the beginning, each data point forms a cluster (also called a node).
 Merge nodes/clusters that have the least distance.
 Go on merging
 Eventually all nodes belong to one cluster

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 71

Clustering Techniques
Hierarchical Clustering:
Pros
- Dendograms are great for visualization
- Provides hierarchical relations between clusters
- Shown to be able to capture concentric clusters
Cons
- Not easy to define levels for clusters
- Experiments showed that other clustering techniques outperform hierarchical clustering

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 72

36
 10-09-2025

Clustering Techniques
Probabilistic Clustering:
 Probabilistic model based clustering assumes that a cluster is a parameterized
distribution.
 Using the data to be clustered as the observed samples, can estimate the
parameters of the clusters.
 A mixture model assumes that a set of observed objects is a mixture of
instances from multiple probabilistic clusters.

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 73

Need for Clustering Techniques in Decision

Making [Power of Clustering]
 Clustering provides valuable insights that aid strategic decision making across various
industries.
 By grouping similar data points, organizations can identify emerging trends, and
consumer preferences.
 Enhanced Customer segmentation – Identify the distinct groups or segments within
a customer base.
 Optimal Resource Allocation – Identify patterns and similarities in data related to
resource utilization.
 Fraud Detection – Groups of transactions that deviate from normal behavior.

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 74

37
 10-09-2025

Role of Clustering in Embedded Systems

 Clustering algorithms are better suited for more complex processing tasks, such as
organizing large datasets into clusters.
 They are useful for identifying previously undetected patterns in data and can help
identify features useful for categorizing data.

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 75

References
• [Link]
• chrome-
extension://efaidnbmnnnibpcajpcglclefindmkaj/[Link]
anils/ppt/Unsupervised%20Learning_CSE543_ASh.pdf
• [Link]
python/
• [Link]
• [Link]
clustering-from-scratch-in-python-a277c23563ac
• [Link]

9/10/2025 [Link], CPS & SCOPE, VIT-CC, Chennai-127. 76

38
 10-09-2025

THANK YOU

9/10/2025 [Link], Associate Professor, DCSEE, NITTTR, Bhopal, Madhya Pradesh, INDIA-462002. 77

39