Research And Implennmentation On Recognition Method For Phishing Mails Based On Text Analysis

Phishing e-mail attack, the main attacking method of the network phishing, which has been the major safety threat for the internet for a long time, has caused massive financial losses and influenced the development of the internet as well as the electronie commerce. The attacking process is that attackers cheat the username and password of the victim's email-boxes or other accounts by sending edited phishing emails, then carry out further attacking behavior and finally cause a great loss and threat for the victim.The past recognizing methods for phishing emails are mainly based on the black&white list, website link, sight similarity and so on. In this paper, we design a new method based on the text analysis to distinguish whether an email is a phishing mail or not. It can be added into the mailbox and tied with the inbox to judge every single received mail. We summarize the features of the common emails as well as the phishing mails we collected. When there is a new-received e-mail we get the features of it. Then we compare the new-received email 's features to the features of the common mails together with the phishing mails we already got with the associative classification method and finally get the new-received e-mail's ownership, whether the new-received email is a phishing email, or really written by the mail sender.