International Conference on Computer and Knowledge Engineering

Home / 12th International Conference on Computer and Knowledge Engineering

Cross-project Defect Prediction with An Enhanced Transfer Boosting Algorithm

Authors :

Nazgol Nikravesh¹ Mohammad Reza Keyvanpour²

1- Data Mining Laboratory, Department of Computer Engineering, Faculty of Engineering, , Alzahra University, Tehran, Iran 2- Department of Computer Engineering, Faculty of Engineering, Alzahra University, Tehran, Iran

Keywords :

class imbalance،software defect prediction،cross-project defect prediction،transfer learning،training data selection

Abstract :

A growing number of software projects makes it increasingly crucial to predict software defects. If sufficient historical data is available, within-project defect prediction models can be effective. During the early stages of software development, however, insufficient data exists to train an effective predictor. Cross-project defect prediction (CPDP) uses information from previous mature projects (source data) to predict whether new software modules (target data) will be defective. CPDP models must take into account the fact that data distributions between target and source projects are different. These models often reduce distribution differences by either selecting training data or using transfer learning methods. Using transfer learning effectively reduces distribution differences in recent CPDP models. Yet none of them have taken into account the possibility that negative transfer may occur as a result of imbalanced nature of defect data. In this paper, a four-step model is proposed, of which three steps are dedicated to the preparation of training data and their initial weights for using in the fourth step, which involves an enhanced version of the transfer boosting algorithm. In this algorithm imbalance nature of data is considered and the weighting of the source data is updated to enhance the prediction performance. Therefore, aside from reducing distribution differences between source and target data, the model also addresses issues related to defect data class imbalance. As compared to four state-of-the-art CPDP models, this model provided consistent and accurate predictions for fifteen projects from PROMISE, AEEEM, and SOFTLAB. Our proposed model provided the best average results for both AUC and F-measure and in some datasets, the improvements were more than 5%.