Threat Eye - An Application of Safety Alert System for Real Time Threat Detection

Tejas S, Shilpa R

Abstract: The "Emergency Threat Detection Android Application" addresses the critical need for swift and reliable emergency response by leveraging advancements in AI and mobile technology. Using Natural Language Processing (NLP), the app analyses user inputs, such as text and speech, to detect emergencies in real time by identifying distress keywords like "help me," "fire," or "accident." Upon detection, it triggers automated actions, including notifying emergency contacts, sending location-based alerts, and providing assistance options. The app features a user-friendly interface, speech-to-text conversion, and a supervised NLP classifier trained on emergency data for precise threat identification. It integrates GPS for real-time location tracking and cloud-based APIs for advanced functionalities like messaging and speech recognition. Designed with scalability in mind, the app supports future enhancements such as multilingual capabilities, smart device integration, and enhanced privacy measures. By reducing response times and improving outcomes during crises, this project demonstrates the potential of combining NLP and mobile technologies to enhance personal safety and emergency management.

Keywords: Safety alert system, Real-time threat detection, Speech recognition, Location tracking, Real-time image capturing