Software Security Testing Model Based on Data Mining

Abstract

Since computers have entered into various fields of daily social life, the reliability and security of computer systems are of significant importance to various fields. The discovery of software security defects is a step-by-step exploration process. Summarizing software security defects is a dynamic process. With the continuous emergence of new software technologies, typical software security defects in different periods may change. The purpose of this paper is to study the software security testing model based on the technology of data mining. Based on the software security testing model foresaid, a testing framework is proposed to guide software security testing. First and foremost, analyze the software interface to obtain static information, for example, parameters; In the second place, dig out the specifications of software requirements according to the obtained parameters, and get efficient component methods and post conditional sequence set. After that, create a set of test cases to test the project and dynamically monitor the information generated by the test, then get the trace set, analyze the trace set, and finally get the security vulnerabilities of the project. After two sets of use-case tests, we can see that the system has found all existing defects in the second regression test.