Automatically Generating Classifier for Phishing Email Prediction

Phishing is a form of online identity theft that employs both social engineering and technical subterfuge to steal consumers´ personal identity data and financial account credentials. Phishing email prediction has drawn a lot of attention from many researchers. According to current anti-phishing research, a classifier generated by decision tree produces the most accurate predictions. However, there appears not to be any open source available to transfer such a decision to an implementable classifier. The work presented in this paper builds a decision tree parser which automatically translates a decision tree into an implementable program language so that the decision is useful in real world applications. Experiment results show that the parser performs as well as the original decision.