Using active learning selection approach for cross-project software defect prediction

Cross-project defect prediction (CPDP) technology can effectively ensure software quality, which plays an important role in engineering. When encountering a newly developed project with insufficient training data, CPDP be used to build predictors using other projects. However, does not take into account the prior knowledge of target items and class imbalance source item data. In this paper, we design active learning selection algorithm for cross-project alleviate above problems. First, use clustering algorithms filter label some representative data from these as guide items. Then, is Finally, balanced cross-item dataset constructed algorithm, model built. article, selected 10 open-source projects by common models, algorithms, evaluation metrics. The results show that proposed solve problem improve performance.