Integrated Machine Learning Model for an URL Phishing Detection

Abstract

The problem of phishing attacks in enterprise is next issue rising in wide scale and complexity, as phishers use email phishing via obfuscated, malicious or phished URLs and continuously adapt or innovate their strategies to lure victims. To gain trust and confidence of victim’s phishers have started using visceral factors and Familiarity cues. Although in most cases a phisher’s clear motive is to commit identity theft in order to benefit from it financially; it is wrong to assume that phishing is always money centric. A phisher can also rob an internet user of his goodwill and character. There are no limits to what a phisher can do in such a scenario. Earning a bad name for oneself in a professional or academic arena can prove much more traumatic than being embarrassed at a social networking site. It is a challenging task to address this issue. It is evident through extensive literature review that single phishing detection filter approaches are insufficient to detect different categories of phishing attempts in enterprise environ. Therefore, a novel anti-phishing model for enterprise using artificial neural network is proposed. In addition, this model effectively identifies whether the phishing email is known phishing or unknown phishing to reduce the trust and familiarity-based email phishing enterprise environ. The Feed-Forward Backpropagation and Levenberg- Marquart methods of Artificial Neural Network (ANN) are adopted to enhance the URL classification process and with Fuzzy Inference System to get result with imprecise data of social features. The proposed model can accurately classify the known and unknown email phishing via URLs.