Behavioral Biometrics for Continuous Authentication

Traditional login techniques passwords or one-time biometrics only verify a user at the start of a session, leaving systems vulnerable to misuse if someone else takes over afterward. Continuous authentication helps solve this problem by regularly checking if the person using the device is still the authorized user. This paper introduces a system that uses behavioral biometrics such as keystroke dynamics patterns, mouse movements, and voice recognition to monitor user identity in real time. These behaviors are unique to each individual and are difficult for an impostor to mimic. Using machine learning, the system learns these patterns and quickly detects when behavior doesn’t match the original user. It works silently in the background without interrupting the user experience. Tests on various users show that the system is both accurate and reliable, with low chances of false alarms. This makes it a strong option for improving session security in everyday digital use.