Comparative study of machine learning test case prioritization for continuous integration testing

There is a growing body of research indicating the potential machine learning to tackle complex software testing challenges. One such challenge pertains continuous integration testing, which highly time-constrained, and generates large amount data coming from iterative code commits test runs. In setting, we can use plentiful for training predictors identify cases able speed up detection regression bugs introduced during integration. However, different models have fault prediction performance depending on context parameters example, variable time budget available cycles, or size execution history used prioritize failing cases. Existing studies case prioritization rarely study both these factors, are essential practice. this study, perform comprehensive comparison approaches that shown best tasks in literature. We evaluate accuracy classifiers predicting fault-detecting tests values with lengths classifiers. evaluation, real-world augmented industrial datasets The results show model budgets execution. Our imply should be carefully configured achieve optimal performance.