摘要
对于程序功能场景的识别一直是程序理解方面的难点之一,现有的方法主要还是依靠人工标注。针对该难点,对移动应用的执行过程进行研究,提出一种基于模型的功能场景自动化标注方法AFSLM。基于交互流建模语言IFML国际标准构造刻画移动应用执行特征的应用执行路径ARP动态模型,用于直接对应移动应用的测试脚本,设计用于度量功能场景相似度的模型匹配算法,提出基于模型的功能场景自动化标注方法AFSLM,实现将人工测试模型的功能场景标签泛化到自动化工具探索模型上,实现高效的功能场景自动化标注。通过实例研究和实验评估,展示方法的有效性。
The recognition of program function scenes is always one of the difficulties in program understanding.The existing methods mainly rely on manual labeling.In view of this difficulty,the implementation process of mobile applications was stu-died,and a model-based automatic labeling method for functional scenes,AFSLM,was proposed.Based on the IFML international standard,an ARP dynamic model was constructed to describe the execution characteristics of mobile applications,which was used to directly correspond to the test scripts of mobile applications,and a model matching algorithm was designed to mea-sure the similarity of function scenarios.AFSLM was proposed to generalize the function scenario labels of manual test models to the automatic tool exploration model,efficient automatic labeling of functional scenarios was realized.A case study and the experimental evaluation were carried out to illustrate the effectiveness of the method.
作者
郑晓梅
钱正轩
李刚
王天舒
ZHENG Xiao-mei;QIAN Zheng-xuan;LI Gang;WANG Tian-shu(School of Artificial Intelligence and Information Technology,Nanjing University of Chinese Medicine,Nanjing 210023,China;Department of Computer Science and Technology,Nanjing University,Nanjing 210023,China;State Key Laboratory for Novel Software Technology,Nanjing University,Nanjing 210023,China)
出处
《计算机工程与设计》
北大核心
2023年第10期3039-3046,共8页
Computer Engineering and Design
基金
国家自然科学基金项目(61472180)
江苏省社会科学基金项目(21TQB001)
江苏省科技计划项目青年基金项目(BK20210696)
未来网络科研基金项目(FNSRFP-2021-ZD-24)。