Crime Awareness and Safety System for Travelers

Abstract

D. Chandan Lagubigi, Kiran Raj K M, Gururagavendra Paluri, Harish Kunder, Ganeshraj S and Sathivik S

Travelers today face a complex environment of threats—ranging from street crime and confidence scams in urban areas to violent crime and natural disaster in rural areas. This paper presents the design, deployment, and evalu- ation of a ”Crime Awareness and Safety System for Travelers” (CAST), an end-to-end mobile system that combines real-time data fusion, ma- chine-learning–based risk prediction, and context- aware alerting to enhance traveler safety. CAST collects multi- source inputs—such as official crime reports, social media trends, geotagged incident reports, and weather data—and fuses them through a multi-layered analytics pipeline. A supervised learning model classifies incoming events by risk and probability, and a spatiotemporal risk map continuously shows hotspots and emerging threats. Users’ feedback highlighted the system’s ease of use and robustness of alerts, although issues remain in ensuring data privacy and avoiding false positives. We conclude that CAST is a scalable solution to traveler safety, with potential extensions to include peer-to-peer reporting and blockchain- based verification of incident data.

PDF