Classification of phishing websites using extreme learning machine and hybrid bat algorithm

Phishing is an electronic fraud through which an attacker can gain access to user credentials. Phishing websites are the one which mimics the legitimate websites and fraudsters evade their detection without much effort. The effect of phishing attack raises the necessity of anti-phishing mechanisms. Several approaches are there to recognize phishing websites such as whitelist, blacklist, machine learning and heuristic-based approach. This paper investigates extreme learning machine and hybrid bat algorithm for the classification of website phishing attacks. The features that distinguish the legitimate from the fraudulent website is selected using the random forest algorithm. The comparison of these techniques with other classification methods, for instance, support vector machine, logistic regression, decision tree and the random forest is also carried out. The results of the classification accuracy show that the hybrid bat algorithm achieves better classification accuracy compared to other classification methods.