🤖 Machine Learning Approaches

AKTU BCS-055 — Unit I

Artificial Neural Network • Clustering • Reinforcement Learning • Decision Tree • Bayesian Network • SVM • Genetic Algorithm

📌 Introduction

Machine Learning can be implemented using different approaches depending on the
type of problem, available data, and desired output.

A Machine Learning Approach is a method or technique used by a machine
learning system to learn patterns from data and make predictions or decisions.

1️⃣ Artificial Neural Network (ANN)

An Artificial Neural Network is a computational model inspired by the
structure and functioning of biological neurons in the human brain.

🔹 Basic Structure

Each neuron receives inputs, applies weights and an activation function, and
produces an output.

$$
z = \sum_{i=1}^{n}w_i x_i+b
$$

$$
y=f(z)
$$

2️⃣ Clustering

Clustering is an unsupervised learning approach in which similar data
objects are grouped into clusters.

$$
\boxed{\text{Similar Objects} \rightarrow \text{Same Cluster}}
$$

🔹 Example

A clustering algorithm automatically discovers groups without requiring
predefined class labels.

📌 Popular Algorithm

$$
\boxed{\text{K-Means Clustering}}
$$

In K-Means, data points are assigned to the nearest centroid and the centroids
are repeatedly updated.

3️⃣ Reinforcement Learning

Reinforcement Learning is a learning approach in which an agent interacts
with an environment and learns from rewards and penalties.

$$
\boxed{
\text{Agent}
\xrightarrow{\text{Action}}
\text{Environment}
\xrightarrow{\text{Reward}}
\text{Agent}
}
$$

📌 Example

A robot learns how to reach a destination. A successful action receives a
positive reward, while an undesirable action may receive a penalty.

4️⃣ Decision Tree Learning

A Decision Tree is a supervised learning approach that makes predictions by
applying a sequence of decision rules.

🔹 Components

• 🌱 Root Node — first decision.
• 🌿 Internal Nodes — additional decisions.
• 🌳 Branches — outcomes of decisions.
• 🍃 Leaf Nodes — final prediction.

5️⃣ Bayesian Networks

A Bayesian Network is a probabilistic graphical model that represents
relationships between variables using a directed acyclic graph (DAG).

📌 Bayes’ Theorem

$$
\boxed{
P(A|B)=
\frac{P(B|A)P(A)}
{P(B)}
}
$$

Bayesian Networks are useful when uncertainty and probabilistic relationships
are important.

6️⃣ Support Vector Machine (SVM)

Support Vector Machine is a supervised learning algorithm mainly used for
classification. It finds a separating hyperplane that maximizes the margin
between classes.

📌 Maximum Margin

$$
\boxed{
\text{Best Hyperplane}
======================

\text{Maximum Separation Margin}
}
$$

For a linear classifier:

$$
\boxed{
w^Tx+b=0
}
$$

The decision rule can be written as:

$$
f(x)=
\operatorname{sign}(w^Tx+b)
$$

⭐ Support Vectors

The data points closest to the separating hyperplane are called
support vectors. They play an important role in determining the optimal
hyperplane.

7️⃣ Genetic Algorithm

A Genetic Algorithm is an optimization technique inspired by the process of
natural evolution.

🔹 Chromosome Representation

$$
101101
$$

Each chromosome represents a possible solution.

🔹 Crossover

Suppose two parent chromosomes are:

$$
P_1=101|110
$$

$$
P_2=110|001
$$

After crossover:

$$
C_1=101|001
$$

$$
C_2=110|110
$$

🔹 Mutation

A small random change may occur:

$$
101101
\quad\longrightarrow\quad
101001
$$

Thus, the algorithm searches for better solutions over successive generations.

📊 Quick Comparison — AKTU Exam Point of View

🎯 AKTU BCS-055 Exam Point

Question: Explain the different approaches of Machine Learning.

Answer: Machine Learning can use several approaches to learn patterns
and make decisions. Important approaches include:

1. Artificial Neural Network
2. Clustering
3. Reinforcement Learning
4. Decision Tree Learning
5. Bayesian Networks
6. Support Vector Machine
7. Genetic Algorithm

For an examination answer, write the definition, working principle,
important components, and one suitable example of each approach.

🧠 Quick Memory Trick

🧠 ANN → Neurons
🔵 Clustering → Groups
🤖 Reinforcement → Rewards
🌳 Decision Tree → Decisions
🎲 Bayesian → Probability
📏 SVM → Maximum Margin
🧬 Genetic Algorithm → Evolution

🚀 PostNetwork Academy

Machine Learning • Artificial Intelligence • Data Science

AKTU BCS-055 — Machine Learning Notes

Learn → Visualize → Practice → Master