Intelligent Local Search for Test Case Minimization
- PDF / 1,399,192 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 61 Downloads / 243 Views
ORIGINAL CONTRIBUTION
Intelligent Local Search for Test Case Minimization Sudhir Kumar Mohapatra1
•
Arnab Kumar Mishra2 • Srinivas Prasad3
Received: 17 October 2018 / Accepted: 15 August 2020 The Institution of Engineers (India) 2020
Abstract For performing efficient regression testing, minimization of test suites is one of the primary approaches. Various kinds of test case minimization techniques have been proposed in the past, in order to do this minimization. However, due to the inherent hardness of this problem, the search for an efficient approach is still going on. In this paper, we propose the application of an intelligent local search algorithm (STAGE), for doing this optimization. The proposed approach performs local search with multiple restarts, using Hill Climbing. But the restart points for the local search are not chosen randomly, rather intelligent decisions are taken for choosing the next starting point. We have observed promising results for the selected subject programs, upon the application of this approach. Keywords Regression testing Test case minimization Intelligent local search Representative set Test case reduction
& Sudhir Kumar Mohapatra [email protected] 1
Department of CSE, IT and SE, Addis Ababa Science and Technology University, 16417 Addis Ababa, Ethiopia
2
Department of Computer Science and Engineering, National Institute of Technology Silchar, Silchar, Cachar, Assam 788010, India
3
Department of Computer Science and Engineering, GITAM University, Visakhapatnam, Andhra Pradesh 530045, India
Introduction Every time new functionalities are added to a software, new test cases are also added in order to test the performance of the newly added functionalities. The different versions of a software go through multiple testing and retesting in the phase of regression testing. As a result of this kind of multiple testing and retesting, redundant test cases get added to the complete test suite. This happens because many times, multiple test cases cover the same part of the code, or in other words, multiple test cases satisfy the same requirements. The process of removing these redundant test cases is called test case minimization, and the resultant reduced test suite is called as the representative set [1, 2]. Given a test suite T = {t1, t2, t3, …, tn}, and the requirements to be satisfied by the suite, R = {r1, r2, r3, ….,rm}, the test case minimization problem is to find the smallest representative set T0 B T, such that T0 satisfies all the requirements in R, which is also satisfied by T. This problem of finding T is unfortunately NP-Complete. A straightforward reduction to the set-cover problem can be shown to prove this [3, 4]. Since the optimization problem is inherently one of the harder problems out there, many approximate optimization algorithms have been applied in the past [5, 6]. In this paper, we propose the application of an algorithm that works as a state-space search algorithm on a complete output space. The proposed approach makes use of a local
Data Loading...
