Smart Crime Analytics: City-Level Cybercrime Prediction Using KNN

This research proposes a machine learning-based system for predicting cybercrime rates across Indian cities using the K-Nearest Neighbors (KNN) algorithm. The study leverages historical crime data, including solved and unsolved cases, population metrics, and socio-economic indicators, to forecast regional cybercrime trends. The system classifies cities into high, moderate, or low-risk zones and offers visual analytics through maps and dashboards. The platform supports real-time interaction, allowing users to explore crime patterns dynamically and identify areas of concern. Additionally, the project includes a social media fake account detection module for enhanced digital security. The backend is implemented in PHP with MySQL, and the frontend uses HTML, CSS, JavaScript, and Google Maps API. The results demonstrate the model's effectiveness in aiding law enforcement and policymakers in crime prevention and resource planning.