Fuzzy Logic Framework Detects Crime Hotspots Using Community Reports

A new research paper from Ariton Verush and Vaibhav Motwani introduces a fuzzy logic framework for community-aware crime hotspot detection, designed as part of an Urban Computing seminar project. The system combines citizen-submitted reports, historical crime data, and contextual urban indicators (e.g., lighting, foot traffic) into a fuzzy inference engine that outputs localized risk levels on a spectrum rather than a binary hot/not-hot classification. This allows the platform to handle uncertainty, partial information, and gradual risk—common in real urban environments. The architecture includes privacy safeguards, multilingual accessibility, and a real-time notification workflow to alert local communities. An exploratory validation with 25 participants measured perceived usefulness, notification relevance, usability, trust, privacy concerns, and acceptance of community reporting. Results indicated that participants felt the platform could improve situational awareness and support reporting and planning discussions, though the study did not measure actual crime reduction or predictive accuracy. The authors present this as a responsible, human-centered prototype for future research, with a PDF and Python prototype available on arXiv (arXiv:2607.16218).