A method and system for automatically detecting phishing for newly added domain names

Abstract

The present invention relates to a method and system for automatically detecting phishing for an added domain name. The method comprises: 1) completely reading a domain name registration database, calculating a domain name increment according to results obtained by reading for successive two times and then expanding the added domain name into an added URL as a to-be-detected data source; 2) obtaining a webpage text content of the added URL as a basis for phishing website identification; 3) representing different parts of the obtained webpage text content as feature vectors; and 4) combining the feature vectors for representing the different parts of the obtained webpage text content into a full text feature vector, calculating the phishing probability of the full text feature vector by using weight information obtained by means of classification algorithm learning, and further generating a list of suspected phishing URLs. The present invention provides a simple and effective method for obtaining an added domain name; the method is applied to automatic detection and identification of the phishing, so that the identification time of a phishing website can be shortened, the detection scope is expanded, and the identification misdetection rate is reduced.

Description

technical field [0001] The invention belongs to the field of network technology and information security technology, and in particular relates to a method and system for automatically detecting phishing for newly added domain names. Background technique [0002] Phishing (phishing) is a fraudulent act of forging online identities. It uses social engineering, technical camouflage and other attack methods to defraud users of their trust. Fake websites that closely resemble real websites to steal sensitive information such as users' personally identifiable data and financial account numbers. With the rapid development of Internet technology and the increasing popularity of e-commerce, more and more commercial activities have begun to conduct online transactions. With the increase of transaction amount, the harm of phishing has become more and more serious. Therefore, the automatic detection and determination of phishing can effectively prevent the disclosure of confidentiality...

AI Technical Summary

Problems solved by technology

[0005] On the other hand, the currently popular phishing website identification method is based on heuristic learning technology, which identifies phishing attacks by analyzing the characteristics of web page content. This method has high accuracy, but it cannot guarantee the comprehensiveness of detection. It is easy to miss the judgment. To solve this problem, the present invention proposes a weighting method based on the feature position to reduce the missed judgment rate

Images