Abstract

This project focuses on Crime Rate Prediction and Safety Navigation using Google Maps, aiming to enhance public safety in urban environments. The system integrates machine learning techniques with spatial data from the Google Maps API to identify crime-prone regions and suggest safer travel routes. Historical crime records, combined with environmental and traffic-related features such as population density, road conditions, street lighting, and accident hotspots, are used to train predictive models including Support Vector Machine (SVM), Random Forest, and Neural Networks. Each area is classified into different crime risk levels, which are then visualized through heatmaps and integrated into Google Maps routing services. Instead of relying solely on the shortest or fastest route, the system provides users with alternative navigation paths that minimize exposure to high-crime areas. Experimental results demonstrate that the approach successfully reduces the probability of traveling through unsafe regions while maintaining reasonable travel times. This work highlights the potential of combining crime analytics with navigation systems to support crime prevention strategies and empower individuals with safer mobility choices. Future enhancements include incorporating real-time data, applying advanced deep learning models, and expanding the system for deployment across multiple cities.