This C++ from-scratch project implements a machine learning system to classify images of washers and coins using the K-nearest neighbors (Knn) classifier and K-means clustering for segmentation. The system incorporates Sobel edge detection and Hu moments for shape analysis, allowing it to accurately distinguish between similar circular objects.

Unsupervised clustering and PCA analysis of African-origin compounds using KMeans (k=4). The project identifies molecular feature patterns, visualizes clusters in reduced PCA space, and extracts representative compounds for each group.

I have applied KMeans, Hierarichal, DBSCAN Clustering on the dataset and shown which Cluster is highly bounded