基于软件行为分析的错误定位方法  

Fault localization method based on software behavior

作　　者：吴江南 吴军华 WU Jiang-nan;WU Jun-hua(College of Computer Science and Technology,Nanjing Tech University,Nanjing 211800,China)

机构地区：[1]南京工业大学计算机科学与技术学院

出　　处：《计算机工程与设计》2019年第10期2891-2894,2925,共5页Computer Engineering and Design

摘　　要：为提高软件自动错误定位的准确性和效率,提出一种基于软件行为分析的错误定位方法。针对复杂度较高的软件系统,将软件的行为作为分析对象,使用静态行为分析技术对具有相似性的类聚类成簇,降低软件的复杂度;使用动态行为分析技术分析软件运行时的函数调用序列,将函数调用序列量化成函数调用频谱信息,作为类可疑度计算的判断依据。将所提方法应用于基准程序,并与同类方法比较可知,该方法有效提高了软件自动化错误定位效果。To improve the accuracy and efficiency of software automatic fault localization,a fault localization method based on software behavior analysis was proposed.Using the behavior of the software as the analysis object,the static behavior analysis technology was used to cluster the similar classes into clusters,which reduced the complexity of software.The dynamic behavior analysis technology was used to analyze the calling function sequence of the software runtime to generate the function suspicious information and class suspiciousness.Applying the proposed method to the benchmark program,and comparing with the similar methods,the method can effectively improve the effects of the software automation fault localization.

关 键 词：软件测试 错误定位 行为分析 潜在语义分析 程序频谱 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]