Detection of phishing webpages using weights computed through genetic algorithm

Phishing, the act of impersonating the legitimate websites is one of the most frequently growing Internet crimes. The purpose of the phishing is to acquire the sensitive information of the users and to use that information for illegal financial gain. In this paper we design and implement an intelligent approach for phishing detection. Our approach utilizes ten features of the webpage and proposes a fitness function for categorizing the webpage. Genetic algorithm is utilized for assigning weights to features. We have collected a dataset of 494 phishing and 450 legitimate URL´s. By using this technique 87% of the sites are correctly classified.