MachineLearning1_Assignment3

K-Nearest Neighbors (KNN) Classification

This repository contains a MATLAB implementation of the K-Nearest Neighbors (KNN) algorithm for supervised classification.

Project Overview

The objective of the project is to implement and evaluate a KNN classifier by assigning class labels based on the nearest training samples in the feature space using the Euclidean distance metric.

Features

• Data loading and preprocessing
• KNN classifier implementation
• Euclidean distance computation
• Prediction based on majority voting
• Hyperparameter analysis (number of neighbors, k)
• Performance evaluation using confusion matrices
• Classification accuracy assessment

Methodology

The implemented workflow consists of:

1. Loading and preprocessing the dataset.
2. Splitting the data into training and test sets.
3. Computing Euclidean distances between samples.
4. Selecting the k nearest neighbors.
5. Assigning class labels through majority voting.
6. Evaluating classifier performance using confusion matrices and accuracy metrics.

Objective

The project aims to study instance-based learning techniques and evaluate the effectiveness of the KNN algorithm for classification problems, analyzing the impact of the number of neighbors on model performance.