Shduv0 / Clustering-Algorithms-From-Scratch-Machine-Learning

Clustering-Algorithms-From-Scratch-Machine-Learning

Clustering Algorithms Implementation and Comparison

Description

This project is aimed at implementing the KMeans, DBSCAN, GMM, and hierarchical clustering algorithms from scratch using Python, as well as utilizing the GMM and hierarchical clustering algorithms provided by the sklearn library. The project uses various datasets, including "Mall Customers", make_blobs, make_moons, and make_circles, to compare and evaluate the performance of the different clustering algorithms. The project was implemented using Colab, and utilizes popular Python libraries such as NumPy, Pandas, Seaborn, and Matplotlib.

Getting Started

Dependencies

The following libraries are required to run the project:

• NumPy
• Pandas
• Seaborn
• Matplotlib
• Sklearn

Installing

To install the required libraries, use the following command:

pip install numpy pandas seaborn matplotlib sklearn

Executing program

To run the project, simply open the Clustering-Algorithms.ipynb file in Colab and follow the instructions provided in the notebook.

Acknowledgments

The project was developed as part of a machine learning course, and was inspired by various online resources and examples. Special thanks to the creators of the "Mall Customers" dataset and the Scikit-learn library for providing the data and implementation of the GMM and hierarchical clustering algorithms.