An extensive analysis of efficient bug prediction configurations

Background: Bug prediction helps developers steer maintenance activities towards the buggy parts of a software. There are many design aspects to a bug predictor, each of which has several options, i.e., software metrics, machine learning model, and response variable. Aims: These design decisions should be judiciously made because an improper choice in any of them might lead to wrong, misleading, or even useless results. We argue that bug prediction con?gurations are intertwined and thus need to be evaluated in their entirety, in contrast to the common practice in the ?eld where each aspect is investigated in isolation. Method: We use a cost-aware evaluation scheme to evaluate 60 di?erent bug prediction con?guration combinations on ?ve open source Java projects. Results:We ?nd out that the best choices for building a cost-e?ective bug predictor are change metrics mixed with source code metrics as independent variables, Random Forest as the machine learning model, and the number of bugs as the response variable. Combining these con?guration options results in the most e?cient bug predictor across all subject systems. Conclusions: We demonstrate a strong evidence for the interplay among bug prediction con?gurations and provide concrete guidelines for researchers and practitioners on how to build and evaluate e?cient bug predictors