Title: Performance of classification using a hybrid distance measure with artificial bee colony algorithm for feature selection in keystroke dynamics

Authors: M. Akila; S. Suresh Kumar

Addresses: Department of Computer Science and Engineering, CSI College of Engineering, Ketti, The Nilgiris, Tamilnadu 643 215, India ' Department of Computer Science and Engineering, Vivekanandha College of Technology for Women, Tiruchengode, Tamilnadu 637 205, India

Abstract: Identifying a user based on the analysis of his/her typing pattern is termed as keystroke dynamics. There are four processes involved in keystroke dynamics. They include feature extraction, feature selection and classification. Initially the statistical measures of feature characteristics such as latency, duration and digraph are computed. Artificial bee colony algorithm is implemented for feature selection. The selected features are classified using a hybrid distance measure. A moderate efficiency was achieved with the obtained results.

Keywords: keystroke dynamics; feature extraction; feature selection; artificial bee colony; ABC algorithm; classification; receiver operating characteristic; ROC curve; distance measures; typing patterns.

DOI: 10.1504/IJCISTUDIES.2013.055227

International Journal of Computational Intelligence Studies, 2013 Vol.2 No.2, pp.187 - 197

Received: 12 Jul 2012
Accepted: 19 Feb 2013

Published online: 19 Jul 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article