A Framework for Multi Features based Phishing Information Identification using NB and SVM Approach

Criminal organizations around the world use the technique known as phishing to extract information from innocent citizens in order to access their bank details, to steal identities, to launder money and more. There are Different types of statistical learning based classification methods are available to differentiate the phishing webpage’s from the original. Feature extraction method is the concept, which has been implementing into the development of web phishing information detection technique. Naïve Bayes and SVM statistical algorithms are used for feature extraction of URL and source code respectively. In contrast to other proposals, this scheme has a high detection rate and a low false negative rate as well as can achieve high detection accuracy, the lower detection time and performance with the small sample of a classification model training set. General Terms Phishing, SVM, Stemming, Web page classification, Feature fusion strategy, Naïve Bayes,URL.