Regresyon testi için test durumu seçim tekniklerinin analizi

Today, the development of software products to satisfy users' requirements has been expedited by the relentless competition that exists in the modern world. These products are in a state of intense competition with one another. Engineers who are developing these products are forced to compete against time as a consequence of this competition. The necessity of developing products quickly within a short period brings along software errors. Testing units bear the responsibility of identifying these software errors before users, as they must quickly execute their pre-prepared test cases. In this regard, testing selected test cases within a limited time has become a necessity. In terms of the Average Cost per Execution (APXC) measure, the goal of this project is to make the regression test process as efficient as possible by making it easier for test engineers to do regression tests. It shows two approaches to testing the right test case in a short amount of time. The first approach named clustered nature-inspired selection aims to conduct testing through selected tests within the test case set. It is based on clustering similar tests within the test case set using Jaccard, a similarity-based approach existing in the literature, and then selecting test cases within the clustered subsets using nature-inspired algorithm approaches. The second approach named direct nature-inspired selection, on the other hand, conducts a selection process directly using nature-inspired algorithms without any grouping among the test cases. These approaches are designed to perform multi-layered selections, discussing which combination among the approaches will yield the optimal result