Combined use of pattern recognition algorithms for keystroke-based continuous authentication system

In this paper we propose several solutions to combine the use of two algorithms for patterns recognition and user validation, both intended to be used in the same time to improve the overall performance of the validation process in a continuous authentication system based on keystroke dynamics. The first one is based on a Multi Layer Perceptron (MLP) Neural Network and the second one relies on a trust algorithm that uses the Bhattacharyya coefficient. For both we present some new evaluation result and conclusions.