Literature Survey on Keystroke Dynamics for User Authentication

Abstract

Behavioural biometrics is the field of study related to the measure of uniquely identifying and measuring the patterns in human activities. Computer security plays a vital role as most of the sensitive data is stored on computers. Keystrokes Dynamics is a technique based on human behaviour for typing the password. Whenever any user logins into the system, username and password combinations are used for authenticating the users. The username is not secret, and the imposter acts as user to guess the password also because of simplicity of password, the system is prone to more attacks. In this case biometrics provide secure and convenient authentication. Our system uses a Support Vector Machine (SVM) which is one of the best known classifications and regression algorithm. Support Vectors (SV) that fall under different regions is separated using hyper planes linear as well as non-linear. Researchers have proved that SVM will converge to the best possible solution in very less time.