摘要
在开放网络环境下软件容易受到攻击,导致软件故障,需要进行安全性测试,针对无监督类测试方法开销较大和复杂度较高的问题,提出一种基于半监督自适应学习算法的软件安全性测试方法;首先采用模糊度量原理构建软件安全测试的半监督学习数学模型,分析软件产生安全性故障的数组特征,然后通过软件故障的熵特征分布方法进行软件的可靠性度量,在开放式网络环境下建立软件可靠性云决策模型,实现安全性测试和故障定位;最后通过仿真实验进行性能验证,结果表明,采用该方法进行软件安全性测试,对软件故障定位的准确度较高,测试的实时性较好,保障了软件的安全可靠运行。
The vulnerable in the open network environment software,lead to software failure,the need for safety testing,the testing method of overhead non supervisory large and complex problems,put forward a kind of software security testing methods based on semi supervised adaptive learning algorithm.First,a semi supervised learning model of fuzzy measure principle construction of software security testing,security feature array fault analysis software,then the software reliability measurement by the method of entropy feature of software fault distribution,the establishment of software reliability of cloud decision model in open network environment,security test and fault location.Finally,through simulation experiments verify the performance,results show that using the method of software security testing of software fault location accuracy,real-time test well,guarantee the safe and reliable operation of the software.
出处
《计算机测量与控制》
2017年第8期5-7,19,共4页
Computer Measurement &Control
基金
西安市科技计划项目(CXY1531WL39)
关键词
开放网络环境
软件
测试
安全
半监督学习
open network environment
software
testing
security
semi supervised learning
