Biometric Authentication using Mouse and Eye Movement Data

This paper presents a novel technique to uniquely identify a user using data gathered from a mouse and the users eyes during computer use. The research involves gathering data from users, extracting user features, and training a Neural Network classifier that can identify them. Previous work has attempted to identify users solely by eye or mouse data alone, but we seek to find out if combining the two kinds of data can produce better results. After going through this process and investigating several Neural Network based classification models we conclude that combining both techniques produces better results and will become practical once the hardware is more widespread.