Severity Prediction for Bug Reports Using Multi-Aspect Features: A Deep Learning Approach

The severity of software bug reports plays an important role in maintaining quality. Many approaches have been proposed to predict the using textual information. In this research, we propose a deep learning framework called MASP that uses convolutional neural networks (CNN) and content-aspect, sentiment-aspect, quality-aspect, reporter-aspect features improve prediction performance. We performed experiments on datasets collected from Eclipse Mozilla. results show model outperforms state-of-the-art CNN terms average Accuracy, Precision, Recall, F1-measure, Matthews Correlation Coefficient (MCC) by 1.83%, 0.46%, 3.23%, 1.72%, 6.61%, respectively.