A Bug Rule Based Technique with Feedback for Classifying Bug Reports

As software programs become increasingly large and complex, it is more important to improve the quality of software maintenance. Many software programs rely on bug reports to correct errors in maintenance activities. Bug tracking systems were developed to guide maintenance activities of software developers. However, due to the excessive number of duplicate bug reports, developers spend much time to identify these bug reports. In this study, in order to save developers´ time in software maintenance, we propose a bug rule based classification technique to categorize bug reports. By utilizing developer feedback mechanism in the technique, it distinguishes duplicate and valid bug reports and is expected to improve the accuracy of bug reports retrieval. Finally, we show the feasibility of this technique in experiment and case study.