ENCRYPTED COMMUNICATION WITH INTELLIGENT THREAT DETECTION: A SECURE CHAT FRAMEWORK

The Secure Chat Application is a web-based encrypted messaging platform designed to facilitate confidential communication within enterprises. To further enhance the reliability and security of the system, this work integrates advanced machine learning and signal processing techniques, specifically Long Short-Term Memory (LSTM) networks and the Kalman Filter. The LSTM network is utilized to model and predict user behavior and message patterns over time, allowing the system to detect anomalies such as unauthorized access attempts, message injection, or abnormal activity sequences. These predictions enable proactive security responses and reinforce system integrity. Complementing this, the Kalman Filter is employed to smooth real-time data streams—such as authentication logs, message timestamps, and user actions—thereby filtering out noise and improving the accuracy of anomaly detection and user session monitoring.This hybrid approach not only fortifies the chat environment against evolving security threats but also optimizes performance by enabling real-time synchronization and responsive data validation. Combined with Firebase’s secure backend for identity management and message storage, and a React-based frontend for cross-platform accessibility, the application offers a robust, scalable, and intelligent communication solution for modern enterprises. The incorporation of LSTM and Kalman filtering positions the platform as a proactive system capable of learning from and adapting to user behavior, thereby elevating both the user experience and security posture in organizational communication frameworks