Copyright © 2022 Alessio Borgi
PROJECT SCOPE: Deepening in the main Clustering Techniques. An in-depth exploration of clustering algorithms and techniques in machine learning. This project covers a variety of clustering methods, from traditional algorithms like K-Means and DBSCAN to advanced techniques, providing a comprehensive understanding of their applications, strengths, and limitations. Ideal for researchers and practitioners looking to enhance their knowledge of unsupervised learning and data analysis.
PROJECT RESULTS:
- Data collection through the Performance Monitoring Windows Application (i.e., Built-in dataset).
- Number of components choice through Elbow Method and Silhouette Coefficient.
- K-Means Clustering Deepening: Lloyd and Elkan Algorithm. K-Means++ and Naïve Sharding Initialization.
- K-Medians Clustering Deepening.
- K-Medoids Clustering Deepening: PAM, Voronoi Iteration, CLARA and CLARANS Algorithms.
- Mean-Shift Deepening: Object Tracking and Image Segmentation Applications.
- DBSCAN Deepening.
- GMM Deepening.
- Deepening evaluation Silhouette Score, Accuracy, Purity, Rand Index, Adjusted Rand Index, Davies-Bouldin Index.
- Final Decision Analysis of the best algorithm given my collected data.
PROJECT REPOSITORY: https://github.com/alessioborgi/Deep_Dive_into_the_Clustering_World