Dynamic Analysis of Software Systems Using Execution Pattern Mining

Software system analysis for extracting system functionality remains as a major problem in the reverse engineering literature and the early approaches mainly rely on static properties of software. In this paper, we propose a novel technique for dynamic analysis of software systems to identify the implementation of the software features that are specified through a number of feature-specific task scenar-ios. The execution of task scenarios and application of data mining algorithm sequential pattern discovery on the generated traces allow us to extract common functionality associated with the corresponding feature-specific task scenar-ios. The extracted patterns are used to identify the groups of core functions that implement software features. The proposed approach can be used for program comprehension andfeature to source code assignment. A case study on the Unix Xfig drawing tool has been provided.