MULTI OBJECTIVE TEST CASE MINIMIZATION COLLABORATED WITH CLUSTERING AND MINIMAL HITTING SET
R. Beena, S. Sarala · 2014
Software testing aspires to explore and validate the attribute and potentiality of a program to authenticate and cross-verify the requisite results obtained. The broader bifurcation of testing is Precision Testing and Reliability Testing. Regression testing is part of reliability testing as it testifies the changes or modifications pursued to the software have not curtailed the functionality of the software by introducing any bugs. It is a kind of quality assurance to the modifications carried out. The pivotal role of regression testing is comprehended whenever modification towards development of software takes place. Reexecution of large test suites is perhaps an enigma many a times due to the paucity of resources. Here arises the need for a novel technique to minimize the test suite in order to remove the redundant test cases. With this focus to provide an innovate and time-effective strategy to remove the redundant test cases, this paper presents a multi-objective test suite minimization by considering maximum statement coverage and minimum execution time. This article also concentrates on incorporating a multi objective minimization technique using clustering approach and minimal hitting set. Here, the identification of appropriate clusters is achieved, through the weighted distance function for mixed variable type and the minimal hitting set is obtained using HS_DAG (Hitting Set Directed Acyclic Graph) algorithm. The results of this experiment exhibit that the algorithm proposed works with adequate efficacy in minimizing the test cases.