Publication

Abstract : Management of the First Information Reports (FIRs) has been a chronic problem in law enforcement because of the manual files, slowness in addressing the matters, and inadequate accessibility among citizens. The paper introduces a smart, talking Police Department Chatbot of FIR Complaints Management, which will automate and simplify the complaint filing, tracking, and management procedures. The suggested framework combines Natural Language Processing (NLP), intent detection, and conversation management based on the context to understand user requests in various languages and help them at the guided complaint registration. A safe cloud-based server and encrypted databases guarantee the integrity of data and its traceability between the police jurisdictions. This system is intended to replace traditional web forms or fixed portals with a hybrid deep learning architecture that incorporates Bidirectional LSTM and transformer-based attention layers to detect the intent more accurately and the entity more precisely, and the system is found to be over 96.4 percent accurate on benchmark conversational datasets. In addition, an officer dashboard application that has real-time analytics will enable setting of cases by priority and verification of digital signatures which result in minimal manual overhead and response time. The study adds a scalable, privacy-preserving conversational model, which can be applied to e-governance and law-enforcement automation, which shows quantifiable improvements in access, operational efficiency, and citizen trust.