In recent years, crime prediction has garnered significant interest in ensuring public safety and optimizing resource allocations. A strategy to tackle this challenge is the identification of crime Hotspot, which are regions with higher than average crime rates. Clustering algorithms play a vital role in areas where crime is a problem by classifying related crime episodes according to spatial and temporal characteristics. In this study we implemented and analyzed K-means and Kmedoids clustering algorithm on a large dataset of crime incidents. The optimal number of clusters (k) was determined to be 14 using the silhouette score. Based on the quality of the clustering findings, the algorithms’ performance was evaluated using their silhouette scores, and both algorithms received a score of 1. The results indicate that both K-means and Kmedoids formed well-defined clusters with excellent demonstrating their effectiveness in identifying high crime zone.