Analysis of Crime Data Visualization and Clustering: PCA + K-Means vs. Feature Extraction

This research paper presents a novel approach for crime data analysis and visualization through the integration of Principal Component Analysis (PCA), K-Means clustering, and feature extraction techniques. Leveraging the inherent structure of crime datasets, an enhanced methodology is proposed that combines dimensionality reduction using PCA, followed by K-Means clustering to identify crime patterns. Additionally, feature extraction is used to provide a comparative study to understand which method is better for clustering which makes use of Silhouette scores The approach not only provides insightful crime data visualization but also identifies distinct crime clusters, aiding law enforcement agencies in efficient resource allocation and crime prevention strategies. Experimental results validate the effectiveness of the proposed method in uncovering intricate crime patterns within diverse datasets.