Detecting phishing websites using rule-based classification algorithm: a comparison

Abstract

In today’s time, phishy website detection is one of the important challenges in the field of information security due to the large numbers of online transactions going through over the websites. Website phishing means stealing one’s personal information over the Internet such as system backup data, user login credentials, bank account details or other security information. Phishing means creation of phishy or fake websites which look like legitimate ones. In this research paper, we use the associative classification data mining approach that is also named as rule-based classification technique by which we can detect a phishy website and thereby identifying the better detection algorithm which has a higher accuracy detection rate. The algorithms used are Naïve Bayes and PART algorithms of associative classification data mining approach. Moreover, we classify the websites into a legitimate website or a phishy website from the collected datasets of websites. The implementation will be done on the datasets of 1,353 websites which contain phishy sites as well as legitimate sites. At the end, results will show us the higher accuracy detection rate algorithm, which will more correctly identify phishing or legitimate websites.