An Enhancement of Multi-Factor Weighted Approach Technique in Prioritizing Test Cases by Comparing Similarity Distance

Abstract

Software testing is one of the most critical phases in the software development life cycle model (SDLC), where the quality of a software product is evaluated. Test case prioritization (TCP) is used to prioritize and schedule test case execution to conduct higher-priority test cases to optimize the software testing process. Traditionally, techniques rely on source code or a specification for the tested system. Therefore, numerous factors and techniques have been used to optimize the prioritization process. One of the factors is distance. String Distance aims to find the degree of similarity between the test cases, which helps prioritize the test case according to the dissimilar value. The higher the dissimilarity value, the higher the probability of detecting new faults. Previous research has used Jaccard Distance to measure the distance to prioritize test cases with the same priority value. In the meantime, the Manhattan Distance is used in this research as it provides a better measure of distance. Our aim of this research is to compare and evaluate both Jaccard and Manhattan Distance algorithms in terms of their effectiveness to formulate the enhancement of the previous multi-factor weighted Approach. The research experiment has shown the process of calculating the Distance matrix for the sample Java Programs and subsequent evaluation using the mutation testing approach and APFD calculation. The results of The Average Percentage of Fault Detected (APFD) of the Test case prioritization by the Manhattan Distance matrix have obtained a higher value, validating its hypothesized effectiveness.