Research on regression test method based on multiple UML graphic models

Qu, Mingcheng; Wu, Xianghu; Tao, Yongchao; Wang, Guan‐Nan; Dong, Ziyu

doi:10.1504/ijguc.2020.108444

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The selection of test cases is performed at the design level using the UML model. Qu et al [6], proposed a method using universal UML to handle various graphical model modifications. To obtain a modified UML graphics module structure, regression testing is required, which is determined by domain of influence analysis on the impact of UML modifications on the generated range of graphical model test cases.…”

Section: Introductionmentioning

confidence: 99%

A UML Activity Flow Graph-Based Regression Testing Approach

2023

View full text Add to dashboard Cite

Regression testing is a crucial process that ensures that changes made to a system do not affect existing functionalities. However, there is currently no adequate technique for selecting test cases that consider changes in Unified Modeling Language (UML) activity flow graphs. This paper proposes a novel approach to regression testing of UML diagrams, focusing on healthcare management systems. We provide a formal definition of sequence and activity diagrams and their relationship and construct corresponding activity flow graphs, which are used to develop a regression testing algorithm. The proposed algorithm categorizes test cases into reusable, retestable, obsolete, and newly generated categories by comparing old and new versions of UML activity flow graphs. The methodology is evaluated using a custom-designed hospital management system website as the test case, and the results demonstrate a significant reduction in time and resources required for regression testing. Our study provides valuable insights into the application of UML diagrams and activity flow graphs in regression testing, making it an important contribution to software testing research.

show abstract

Section: Introductionmentioning

confidence: 99%