A background path blasting method based on machine learning

Abstract

The invention relates to a background path cracking method based on machine learning. The method comprises the steps that firstly, URL paths of ordinary websites with background characteristics are crawled, an ordinary background dictionary is generated; secondly, the URL paths in the ordinary background dictionary and the URL paths in an ordinary non-background dictionary are vectorized; thirdly, the vectorized URL paths are trained through a classification algorithm; fourthly, a page of a target website is crawled, a set of all the URL paths of the target website is obtained, the URL paths with the target website characteristics and the URL paths with the ordinary background characteristics are combined, and a background dictionary with the target website characteristics is generated; fifthly, the generated background dictionary with the target website characteristics is input into the trained classification algorithm so as to conduct identification and classification, and an optimal dictionary is obtained; sixthly, the optimal dictionary is utilized to achieve cracking of the background path of the target website by using a multi-thread cracking technology. By means of the background path cracking method, the efficiency and the success rate of website background cracking can be improved.

Description

technical field [0001] The invention belongs to the field of information technology, and in particular relates to a background path blasting method based on machine learning. Background technique [0002] In the prior art, when penetration testing is performed on the website for a long time, it is often stopped at the inability to use the obtained user or administrator information because the background login interface cannot be found. [0003] Existing background scanning technologies are mainly scanners based on multithreading and large dictionaries. For example, the existing background scanning tools Yujian and Coconut are all based on fixed large dictionaries and multi-threaded scanning. Since existing tools are written based on large dictionaries, they are only useful for most websites with poor security awareness, and cannot pose threats to websites with higher security levels, and cannot detect their real sensitive directories and background interfaces. Contents of...

AI Technical Summary

Problems solved by technology

Since the existing tools are written based on large dictionaries, they are only useful for most websites with poor security awareness, and cannot pose threats to websites with high security levels, and cannot detect their real sensitive directories and background interfaces

Images