This repository contains Python notebook to compare the performance of the K-Means and K-Medoids clustering algorithms using a sample dataset. The purpose of this comparison is to understand the differences in clustering results and computational efficiency between these two popular clustering techniques.
The dataset used for this comparison is stored in a file named Data.xlsx
. It contains four features (A1, A2, A3, A4) for each data point. These features are numerical values representing characteristics of each data point.
- k_means.py: Contains the implementation of the K-Means clustering algorithm.
- k_medoids.py: Contains the implementation of the K-Medoids clustering algorithm.
- visualization.py: Provides functions for visualizing the clustering results.
- Ensure that you have Jupyter Notebook installed and setup on your system.
- If not then open the notebook in google colab and start coding.
After running both clustering algorithms, compare the clustering results visually using the provided visualization functions. Analyze the clusters formed by each algorithm and evaluate their effectiveness based on the dataset characteristics.
- The dataset used in this comparison is for demonstration purposes only.
- Feel free to modify the code or dataset to conduct further experiments and analysis.