Biometrics, device metrics and pseudo metrics in a multifactor authentication with artificial intelligence

The last two decades have seen the unprecedented growth of the Internet and the corresponding increase in online services. Hence secure, interoperable and flexible identity management systems have become a fundamental precondition for tenacity of e-services and to alleviate cybercrime. In this paper, a multifactor authentication system is implemented through a fuser block of an artificial neural network and adaptive neural-fuzzy inference system in an effort to create secure authentication systems. The results from the two techniques are then compared to determine the most effective technique. Entropy from Shannon's information theory is used to develop the identity attributes metrics.