Section A: Definitions and Concepts $(15$ Marks$)$

Answer all questions in this section. Each question carries $5$ marks.

Learning Paradigms:

Explain the differences between supervised learning, unsupervised learning, and reinforcement learning. Provide one example for each type.

$2.$ CNNs:

Define Convolutional Neural Networks $($CNNs$),$ describe their architecture, and explain their importance in image processing. Provide a real$-$world application.

$3.$Data Preprocessing:

Explain the purpose of data preprocessing in machine learning. Discuss two techniques, such as normalization and dimensionality reduction, and their impact on model performance.

Section B: Practical Applications $(25$ Marks$)$

Answer all questions in this section. Each question carries $5$ marks.

Linear Regression Code:

Analyze the following Python code:

$"\text{'}$python

from sklearn.linear$\_$model import LinearRegression

import numpy as np

# Data

X $=$ np$.$array$([[1],[2],[3],[4],[5]])$

$y=$ np$.$array$([1,2,1\mathrm{.}3,3\mathrm{.}75,2\mathrm{.}25])$

# Model

model $=$ LinearRegression$()$

model.fit$(x,y)$

# Prediction

print$(\text{'}$Prediction for input $6$:$\text{'},$ model.predict$([[6]]))$

$...$

Explain each line of the code.

Calculate the predicted value for input $6$ and interpret the result.

Compare Breadth$-$First Search $($BFS$)$ and Depth$-$First Search $($DFS$).$ For each, provide:

A brief explanation of how the algorithm works.

A real$-$world application where it would be most suitable.