Unsupervised Clustering Algorithms

This repository contains implementations of three unsupervised clustering algorithms: DBSCAN (Density-Based Spatial Clustering of Applications with Noise), K-Means, and K-Means EM (Expectation-Maximization). These algorithms are used to group similar data points together without any prior knowledge of the labels or categories.

Table of Contents

• Introduction
• Algorithms
  • DBSCAN
  • K-Means
  • K-Means EM
• Installation
• Usage

Introduction

Clustering is a fundamental task in unsupervised machine learning, where the goal is to group similar objects in such a way that objects in the same group (cluster) are more similar to each other than to those in other groups. This repository provides implementations of three clustering techniques:

1. DBSCAN: A density-based clustering algorithm that can find arbitrarily shaped clusters and is robust to outliers.
2. K-Means: A popular partitioning method that divides data into k clusters by minimizing the variance within each cluster.
3. K-Means EM: An extension of K-Means that uses the Expectation-Maximization (EM) algorithm to estimate the parameters of a mixture of Gaussian distributions.

Algorithms

DBSCAN

• Description: DBSCAN is a clustering algorithm that groups together points that are closely packed together, marking points that are far away as outliers. It works well with clusters of varying shapes and sizes.
• Parameters:
  • eps: The maximum distance between two points to be considered as neighbors.
  • min_samples: The minimum number of points required to form a dense region (i.e., a cluster).
• Advantages:
  • Can find arbitrarily shaped clusters.
  • Does not require specifying the number of clusters.
  • Robust to noise and outliers.
• Disadvantages:
  • Not suitable for datasets with varying density.
  • Performance depends on the choice of eps and min_samples.

K-Means

• Description: K-Means is a centroid-based algorithm that partitions the dataset into k clusters, where each data point belongs to the cluster with the nearest mean (centroid).
• Parameters:
  • k: The number of clusters to form.
  • max_iter: Maximum number of iterations for the algorithm.
  • tol: Tolerance to declare convergence.
• Advantages:
  • Simple and fast.
  • Works well for spherical clusters.
• Disadvantages:
  • Requires the number of clusters (k) to be specified.
  • Sensitive to the initial placement of centroids.
  • Assumes clusters are spherical and evenly sized.

K-Means EM

• Description: K-Means EM is a variant of K-Means that incorporates the Expectation-Maximization algorithm. It models the data as a mixture of Gaussian distributions and iteratively refines the cluster assignments and parameters.
• Parameters:
  • k: The number of clusters to form.
  • max_iter: Maximum number of iterations for the EM algorithm.
  • tol: Tolerance to declare convergence.
• Advantages:
  • Handles overlapping clusters.
  • Provides a probabilistic cluster membership.
• Disadvantages:
  • Requires the number of clusters (k) to be specified.
  • Computationally more expensive than standard K-Means.

Installation

To use the clustering algorithms, clone the repository and install the required dependencies:

git clone https://github.com/yourusername/unsupervised-ML_Clustering.git
cd clustering-algorithms
pip install -r requirements.txt