Method for detecting counterfeit webpage

Abstract

The invention discloses a method for detecting a counterfeit webpage, which relates to the technical field of information safety and is used for improving the detection accuracy and detection efficiency of a phishing attack website. The method comprises the following steps of: converting a webpage to be detected into a webpage image and converting the webpage image from a red-green-blue (RGB) space into a hue-saturation-intensity (HSI) space; partitioning the converted webpage image by a spectral clustering method; extracting the characteristic vector of each sub-image after partitioning; extracting the characteristic vector of the position relation between adjacent sub-images after portioning; combining the characteristic vector of each sub-image with the characteristic vector of the position relation between the adjacent sub-images to form a webpage image characteristic vector and performing dimensionality reduction on the webpage image characteristic vector by a kernel principal component analysis method so as to acquire a characteristic space; training and testing the characteristic space by using a transductive support vector machine classifier; and judging whether the webpage to be detected is the counterfeit webpage or not according to the classification result of the classifier. The method has high adaptability and high detection accuracy in the aspect of the detection of the phishing attack website.

Description

technical field [0001] The invention belongs to the technical field of information security, in particular to a method for detecting counterfeit webpages. Background technique [0002] The rapid development of network applications has made online transactions and e-banking an important business model, and private information such as online user accounts has become extremely important. Phishing is a method of defrauding legitimate users of account passwords, etc. through fake websites. Fraudulent process for sensitive information. Phishing webpage detection technology is the process of detecting counterfeit webpages by identifying the characteristics of webpages. At present, most of the Phishing web page detection technologies are based on the document model, but due to the flexibility of the HTML language and the dynamic nature of web page elements, counterfeiters can make web pages that look the same but have completely different structures. In this regard, using the Phis...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method for detecting counterfeit webpages, which solves the problems of insufficient processing of image webpages, high dimensionality resulting in complex calculations and low similarity matching accuracy due to the current purely document-based Phishing webpage detection technology.

Images