Generating training data for SART-2 keystroke analysis module

The keystroke dynamics analysis is one of the biometric techniques which takes into account behavioral features. The difficulty with the keystroke dynamics information is its poor stability. This article briefly presents a solution used to collect training data and extract features for a user authentication system based on the keystroke dynamics. Then a few approaches which might be applied to authenticate users basing on the keystroke rhythm are presented. These approaches are going to be tested in order to choose an efficient method to be applied as a part of a biometric security system for mobile workstations, which is being created within the framework of the SART-2 project.