Unsupervised machine learning

A comparative study of K-centroid clustering algorithms, including KMeans, CustomKMeans, Fermat-Weber KMedians, and Weiszfeld KMedians, highlighting their performance on separated and non-separated datasets.

Customer-Segmentation---Purchasing-Behavior

BPNN, K-means, K-medoids

Clustered behavioral data into two groups, regardless of gender, and evaluated cluster consistency with gender division using silhouette and Davies-Bouldin scores. Additionally, identified the optimal cluster count using the elbow method and re-evaluated clustering efficacy.