Utilizing Creator Profiles for Predicting Valuable User Enhancement Reports

Abstract

Users of software applications use Issue Tracking Systems (ITSs) to file enhancement reports, which leads to a large quantity of user requests. Indeed, these reports have become an important source for software requirements because they help to continuously improve software applications. Usually, developers and maintainers evaluate them and decide which user reports can be accepted. However, enhancement reports are continuously being raised one after another, which makes this process time-consuming and labor-intensive. Timely handling and implementation of these enhancement reports can effectively improve user satisfaction and product competitiveness. Thus, research has focused on automated methods for predicting which enhancement reports are likely to be approved, to maximum the value derived from user reports. However, reported results of existing approaches are typically not good enough for practical use. In this paper, we propose a novel creator-profile-based method to explore dependencies among enhancements to improve the prediction performance. Firstly, we define the concept of a creator profile, including a general method of how to generate creator profiles from the data set. Then we explain how to employ creator profiles to the problem of enhancement report approval prediction. Finally, we evaluate our approach on 40,551 enhancement reports collected from ITSs. The experimental results indicate that our proposed approach greatly improve on existing state of the art, especially in predicting approved reports. For cross-application prediction, the accuracy is 80.7%, while for non-cross-application prediction, the overall accuracy is 83.6%. That is, with the proposed approach, over 80% of user requests can be automatically identified for exacting valuable user requirements, which significantly reduces labor costs. Replication package is available at: https://anonymous.4open.science/r/approval_prediction-507E