Abstract:
The rapid rise in urban crime and public safety concerns has driven the need for inno vative technological solutions. Traditional incident reporting methods lack real-time data processing and predictive capabilities, leading to delayed responses and inefficient crime management. This project aims to develop an Safety-driven incident reporting and hotspot detection application that integrates machine learning, large language models (LLMs), and data visualization. The application will enable users to report incidents in real-time, predict high-risk zones, and provide community verification fea tures to filter out false reports. By leveraging advanced data-driven approaches, the system seeks to enhance public safety, optimize resource allocation, and assist law enforcement agencies in proactive crime prevention
