Abstract: Aiming at the problem of the vulnerability coverage and artificial defect review in the software vulnerability detection,a method for software vulnerability discovery based on the soft set and multi-attribute comprehensiveness was proposed.Firstly,based on trusted integrated detection tools,an evaluation model of software vulnerability factors was established.Secondly,the soft set was introduced to measure vulnerability factors,then the serious impact on software security was determined through the method of multi-attribute comprehensive integration tools,and the discovery process of software vulnerability was finally completed.Experimental results show that the method has better detection capabilities for vulnerability in different level ,which provides a feasible way for the improvement of software vulnerability detection false positive rate and false negative rate.

Key words: Software vulnerability,Soft set,Attribute set,False negative rate,False positive rate