A Comparative Analysis of DBSCAN, K-Means and Agglomerative Clustering Algorithms for Geospatial Data
Anupam Jain, Khushal Rathi, Yuboraj Ganguly, Ankit Kumar, Yogiraj Anil Bhale · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2025
This study presents a comparative analysis of three popular clustering algorithms, DBSCAN and KMeans, Agglomerative Clustering applied to geospatial data.We focus on their performance based on the silhouette score, examining their ability to identify meaningful clusters in noisy data.Our results show that DBSCAN outperforms KMeans and Agglomerative 9oClustering, achieving a silhouette score of 0.8646 compared to KMeans' 0.8160 and Agglomerative Clustering's 0.8160, highlighting DBSCAN's robustness in identifying clusters with irregular shapes and handling noise.