基于深度学习的跨项目软件缺陷预测

ross-project software defect prediction uses the historical data of one or more source projects to establish the defect prediction model, and then carries on the defect prediction to the target project. At present, there are still some problems in the field of cross-project software defect prediction. Aiming at the problem of semantic absence in the field of cross-project software defect prediction, this paper proposes a convolutional cyclic neural network model based on attention wit. The model using the class balance data obtained through sampling the smote algorithm, and then through word embedding algorithm converts program source node vectors as the input of neural network by convolution with attention wit circulation neural network for feature information, eventually to extract can be semantically related characteristics of migration, build the project defects prediction model. The experimental results show that the cross-project defect prediction model based on deep learning proposed in this paper can alleviate the semantic missing problem in the field of cross-project defect prediction to a certain extent.