Leveraging Machine Learning to Enhance Crime Prediction and Community Safety

Understanding crime patterns is essential for effectively addressing criminal behavior. The study analyzes crime data from Indore, publicly available on the Indore Police website, to generate predictions for the types of crimes likely to occur at specific times and locations within the city. The study employs AI and machine learning algorithms to investigate their efficacy in crime prediction, as these technologies have demonstrated success in audio and visual-based crime detection. Such systems have challenges demonstrating their dependability to legal authorities, mainly when employed for event prevention objectives. The research addresses classification issues by categorizing diverse crime types into generalized classifications, enhancing classification systems' significance and practical utility. This study evaluates various machine learning classifiers to identify the optimal classification approach and demonstrates the impact of our results. The study illustrates the capabilities of an AI-driven crime prediction system and offers ideas for improving crime prevention strategies through future research.