Improving Priority Prediction for Bug Reports with Comment Features

Deciding the priority of processing a bug report is an important issue for the efficiency of software bug resolution. Generally, this decision process is labor-consuming because the triager needs considerable time to decide the priority. To mitigate this problem, an automatic method for predicting the priority of a bug report is needed. In this paper, we propose a novel prediction scheme to improve the prediction performance with comment features. We have conducted empirical experiments to evaluate the proposed scheme with a previous top-ranked scheme eApp. The experimental results show that the proposed approach can effectively improve the prediction performance.