User authentication and monitoring based on mouse behavioral features

With an empirical study of mouse behavioral features using qualitative and quantitative analysis from the physiological layer and the interactive layer, an identification method based on sequential forward greedy selection and SVM was proposed. Specifically, an identity verification experiment, in which 20 participants were involved, showed that the performance of proposed method was encouraging with false acceptance rate (FAR) of 1.67% and false rejection rate (FRR) of 3.68% for user classification. Experimental results show that the proposed method have better performance than conventional classification and recognition methods (BP, RBF, SOM), and also provide a strong evidence for the effectiveness and feasibility of user authentication and monitoring based on mouse activities.