EMERGENCY RESPONSE SYSTEM FOR DISASTER MANAGEMENT WITH MULTILINGUAL AI CHAT SUPPORT

Authors:

Shreya Suryavanshi, Aayush Hatkar, Ganesh Raghtate

Department of Electronics & Telecommunication. VIT, Mumbai

Abstract—With accelerating urbanization and the growing frequency of emergencies such as road accidents, medical crises, and natural disasters, the limitations of traditional voice-based emergency response systems have become increasingly apparent. This paper presents the design and implementation of an artificial intelligence (AI)-driven multilingual emergency response chat intelligence system that serves as a digital first responder. The system enables users to communicate in their native language through a secure chat interface, performs real-time intent classification, delivers context-aware guidance, and visualizes nearby hospitals and police stations on a live map. A centralized analytics dashboard further supports data-driven response optimization. The system is evaluated across three emergency phases—pre-emergency awareness, real-time assistance, and post-emergency follow-up—demonstrating strong intent classification accuracy of approximately 92%, a mean response latency of 2.1 seconds, and improved accessibility for linguistically diverse users across five regional languages.

Keywords—Artificial Intelligence; NLP; Emergency Response; Multilingual Chatbot; Machine Learning; Disaster Management; Live Map Monitoring; Data Analytics