Lookup NU author(s): Steven Shorrock,
Professor Satnam Dlay,
Dr Dave Atkinson
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
The number of people with access to the Internet continues to grow at a phenomenal rate. The potential for e-commerce on the web is huge, but fears about security are holding it back. Authentication of a person's identity is a major concern in the fight against fraud. We present a method of identifying a computer user by using the unique rhythm generated when typing on a keyboard. Three neural networks, the cascade forward neural network (CFNN), probabilistic neural network (PNN) and traditional backpropagation neural network (BPNN) are implemented on two types of data set. Each networks performance and suitability to the task is evaluated. The PNN is seen to outperform the BPNN and a novel implementation based around the CFNN displays a great deal of promise.
Author(s): Shorrock S, Yannopoulos A, Dlay SS, Atkinson DJ
Publication type: Book Chapter
Publication status: Published
Book Title: Advances in Physics, Electronics and Signal Processing Applications
Print publication date: 01/01/2000
Publisher: World Scientific and Engineering Society Press
Library holdings: Search Newcastle University Library for this item