Test Suite Prioritization Based on Optimization Approach Using Reinforcement Learning

Regression testing ensures that modified software code changes have not adversely affected existing modules. The test suite size increases with modification to the based on end-user requirements. executes complete after updates in software. Re-execution of new cases along is costly. scientific community has proposed prioritization techniques for selecting and minimizing minimize cost regression testing. goal maximize fault detection minimum cases. Test minimization reduces by deleting less critical In this study, we present a four-fold methodology reinforcement learning. First, testers’ users’ log datasets are prepared using interaction recording systems android application. Second, learning model used predict highest future reward sequence list from data collected first step. Third, algorithm signifies prioritized suite. Lastly, seeding approach validate results engineering experts. learning-based optimization evaluated through five case study applications. performance evaluation show mechanism performs better than baseline approaches random t-SANT approaches, proving its importance