Novel Approach: Prioritizing Test Cases Based on a Comprehensive Taxonomy of Product Metrics

Abstract

Software product quality is still a significant problem, and testing is essential in ensuring software quality. Unfortunately, testing is hurried and frequently poorly executed because it is costly, time-consuming, and occurs near the end of the software engineering process. To address some of these issues and look at ways to lower software testing costs, this paper proposes prioritizing test cases based on a Comprehensive Taxonomy of Metrics (CTM). We have developed a taxonomy of product complexity metrics based on the product level, including the class, methods, and statements. To assess these metrics, we have developed a tool that utilizes them to target test cases. This tool allows us to identify three sample metric combinations. We conducted a series of experiments across three applications, applying these metric combinations to test case prioritization for each application. To investigate the efficiency of our test case prioritization, we created a large number of mutants using MuJava. Our experiments show that the mutant scores of all prioritized classes are remarkably high, and most of our test suites were able to discover 100 % of mutants.