Skip to content

Latest commit

 

History

History
19 lines (15 loc) · 1.35 KB

README.md

File metadata and controls

19 lines (15 loc) · 1.35 KB
Raw

Clustering Deepening

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