Paper Title :User Identity Prediction By Mouse Gesture Dynamics Through ANN

Author :Abhay A. Jadhav, J. V. Megha

Article Citation :Abhay A. Jadhav ,J. V. Megha , (2015 ) " User Identity Prediction By Mouse Gesture Dynamics Through ANN " , International Journal of Advance Computational Engineering and Networking (IJACEN) , pp. 31-34, Volume-3, Issue-3

Abstract : we propose an approach for the user authorization system during login based on the Signature drawn from mouse movement. The scenario presented here is that the system can successfully and easily identify user behavior based on its behavioral model. Our implemented system uses Artificial Neural Network approach to train user behavior and verify user sign pattern for authentication of user to system. In this biometric scenario we have two parts, In First phase, the user signature is created as per the user’s interaction with mouse while he is doing some activity such as, drawing any alphabet or his signature on canvas application and it gets stored in a database and used for verification purpose. In the second phase we have designed hierarchy, to generate a user signature for the verification purpose with signature stored in database. Our experimental results work on ten user signatures drawn for Authorization of user. Each user has to store five various signature patterns or sign variations and at a time of verification user has to draw his signature. If drawn signature matches with any of them, the user will be treated as Valid or Authorized User to system else fake user. We present the results of several experiments that we conducted to state our observations and suggest guidelines for evaluating future authentication approaches based on mouse Gesture dynamics by ANN. Keywords- Mouse Dynamics, Behavioral Biometrics, Ann, Human Computer Interaction, User Re-Authentication, Anomaly Detection.

Type : Research paper

Published : Volume-3, Issue-3

	|		PDF	|	Viewed - 73	|	Published on 2015-03-03