Keystroke Biometrics with number-pad input using hybridization of adaboost with random forest

Keystroke Biometrics is a new authentication technique to identify legitimate users via their typing behavior, which are in turn derived from the timestamps of key-press and keyrelease events in the keyboard while typing their password. Many researchers have explored this domain, with mixed results, but few have examined the relatively impoverished results for digits only password, so that the input password is from the number-pad portion of the keyboard. In this paper, machine learning technique is adapted for keystroke authentication. The selected classification method is adaboost and random forest. Also, combination of adaboost and Random forest will improve the accuracy of the system. The performance metrics are FAR (False Acceptance Rate) and FRR (False Rejection Rate).