Implementation-of-kNN-algorithm-from-scratch-in-Python

This repository contains a simple implementation of the k-Nearest Neighbors (k-NN) algorithm in Python, a popular machine learning method for classification and regression tasks.

Overview

The k-NN algorithm works by analyzing the distances between data points to make predictions based on the majority vote of the k-nearest neighbors. This implementation includes functions for calculating various distance measures (Euclidean, Manhattan, and cosine distance) and finding the k-nearest neighbors for a given data point.

Features

• Simple implementation of the k-Nearest Neighbors algorithm
• Distance calculation using Euclidean, Manhattan, and cosine distance measures
• Data preprocessing functions for MinMax scaling and standard scaling
• Train and test dataset splitting
• Data visualization through scatter plot matrices

Dependencies

• NumPy
• pandas
• Matplotlib
• scikit-learn (only for loading the example dataset)

Usage

• Clone the repository
• Install the required dependencies
• Import the necessary functions and libraries
• Load your dataset and preprocess it (if necessary)
• Apply the k-Nearest Neighbors algorithm to your dataset