Enhanced password authentication through keystroke typing characteristics

Abstract

Biometric security systems enable more secure authentication methods to access a computer system's resources. This paper presents a biometric security system based on keystroke dynamics that enable hardening or strengthening the password verification process by comparing the captured keystroke dynamics information with the user's templates stored in a template database. The system uses the statistical information about keystroke latencies of users as a measure to differentiate different users. It uses a classifier that resembles a neural network as the structure.