Abstract Context Cross-project Defect Prediction (CPDP) utilizes other finished project (i.e., source project) data to predict defects of the current working project. Transfer Learning (TL) has been mainly applied at CPDP to improve prediction performance by alleviating the data distribution discrepancy between different projects. However, existing TL-based CPDP techniques are not applicable at the unit testing phase since they require the entire historical target project data. As a result, they lose the chance to increase the product’s reliability in the early phase by applying the prediction results.
Objective To increase the product’s reliability in the early phase by proposing a novel TL-based CPDP technique applicable at the unit testing phase (i.e., eCPDP).
Method We utilize Singular Value Decomposition (SVD) that only requires source project data for TL.
Result eCPDP shows similar or better performance than the 5 state-of-the- art TL-based CPDP techniques on 9 different performance metrics over 24 projects.
Conclusion 1) We show that eCPDP is an applicable CPDP model at the unit testing phase. 2) It can help practitioners find and fix defects in an earlier phase than other TL-based CPDP techniques.