/Clustering-Methods-For-Unsupervised-Data

In this project, our goal is to explore various clustering techniques and identify the most suitable method for the given dataset. We implemented K-Means, MiniBatchKMeans, MeanShift, AffinityPropagation, Hierarchical, and DBSCAN, and evaluated their performance using silhouette, Calinski Harabasz, and Davies Bouldin scores.

Primary LanguageJupyter Notebook

Clustering-Methods-For-Unsupervised-Data

In this project, my main objective was to explore clustering techniques without utilizing dimensionality reduction methods. I implemented several clustering algorithms, including K-Means, MiniBatchKMeans, MeanShift, AffinityPropagation, Hierarchical, and DBSCAN, and assessed their performance using metrics like silhouette score, Calinski Harabasz score and Davies Bouldin index. By delving into the raw, high-dimensional nature of the data, we aimed to identify the optimal clustering approach for our dataset, gaining valuable insights into the underlying structures and patterns within the data.