CRIMQA: A Natural Language Question Answering System for Exploring Urban Crime Perceptions and Patterns through Interactive Maps

Davis Coropuna L.F.; Cuadros Valdivia A.M. · Proceedings - International Conference of the Chilean Computer Science Society, SCCC (2025) · DOI: 10.1109/sccc67219.2025.11420807

Resumen

Understanding urban crime and citizen perception is essential for informed decision-making in public safety. However, access to criminological data is often constrained by fragmented sources and technically demanding tools. We present CRIMQA, a system that enables querying crime data through natural language, generating interactive geographic visualizations. Leveraging large language models (LLMs), CRIMQA integrates official records with social media posts, providing contextualized and up-to-date insights. The system supports guided exploration and multilevel spatial analysis, offering transparent explanations that strengthen evidence-based decision-making.

Instituciones: Universidad Nacional de San Agustín de Arequipa