An ensemble algorithm for discovery of malicious web pages

Abstract

Internet has become one of our daily life activities that all of us agree on its important role. It is necessary to know how it can either have misuse. Identity theft, brand reputation damage and loss of customer's confidence in e-commerce and online banking are examples of the damages it can cause. In this paper, we proposed an ensemble learning algorithm for discovery of malicious web pages. The goal is to provide more learning chance to the data instances, which are misclassified by previous classifiers. To this aim, we employ a genetic algorithm (GA) to improve classification accuracy. In this algorithm a weight is assigned to a weak classifier and GA chooses the best set of committee members of weak classifiers to make an optimal ensemble. Experimental results demonstrate that this algorithm leads to the classification accuracy improvement.