Biometrie authentication using keystroke dynamics

Passwords play an important role in the area of security. However though the username and its password is known only to the respective user, this key-value pair can be obtained and used by others. This paper focuses on use of keystroke dynamics as biometric functionality for user identification and verification. As typing pattern of each person is unique, it is calculated based on mainly flight time, dwell time and some other parameters using machine learning techniques. User verification is done using mean and standard deviation method. A threshold value of 70 percent is set to distinguish between a valid and an invalid user.