Fuzzy test case generation method based on machine learning method

Abstract

The invention relates to a fuzzy test case generation method based on a machine learning method, and relates to the field of information security. According to the method, the test case redundancy problem existing in the current fuzzy test technology is optimally designed. In the aspect of generating the fuzzy test case oriented to the source program file, before the fuzzy test case is generated,the stain variable and the problem function in the program object are marked and identified, and the existing seed case generation and screening technology is combined, so that the effectiveness of the fuzzy test case can be improved, and the redundancy of a fuzzy test case set is reduced. In the test case generation link, combining machine learning, the feasibility of machine learning for test case simplification is analyzed, to obtain a test case of machine learning and generate an optimization technique idea. According to the method, a machine learning model and an algorithm are adopted toimprove a test case generation link in a fuzzy test process, improve the generation efficiency of the test case, realize redundancy removal of test case combination and achieve the purpose of improving the intelligent degree of the fuzzy test process.

Description

technical field [0001] The invention relates to the technical field of information security, in particular to a method for generating fuzzy test cases based on a machine learning method. Background technique [0002] According to the statistics of security vulnerabilities released by the National Information Security Vulnerability Database (CNNVD), the number of security vulnerabilities announced in my country in 2018 was 23,029, compared with the total number of security vulnerabilities in 2017, which was 18,586, an annual growth rate of about 23.9%. Compared with the surge in the number of security vulnerabilities in 2017, the growth rate of the number of security vulnerabilities in 2018 slowed down. However, this does not mean that the prevention of security vulnerabilities has achieved gratifying results, because the main reason for this phenomenon is that the statistics and publication of security vulnerabilities are more decentralized than before, and a large number of...

AI Technical Summary

Problems solved by technology

The current mainstream fuzz testing technology in the test case generation stage has serious use case redundancy, high cost, and problems such as uncontrollable cost.

[0006] (3) In terms of type coverage of security vulnerabilities, fuzz testing technology still needs to be improved

Although the current fuzzing technology can effectively discover certain types of security vulnerabilities, it is powerless to some specific security vulnerabilities.

Images