A Survey on Automatic Bug Triage Using Data Mining Concepts

In bug triage process, assigning a correct developer to fix the new incoming bug is tedious than fixing that bug. Since the number of daily new coming bug is high, manual triaging increases the development cost and time. In order to automate the bug triage process, data mining techniques such as text classifications are used. Data reduction is one of the major problem identified in applying text classification to automate the bug triage process, which decreases the accuracy of bug triaging. This paper provides the complete survey of automatic bug triage using data mining concept. Also, a framework called Auto-BugTriager that eliminates the problem of data reduction to greater extent in bug triaging has been proposed. The proposed framework consist of three phases namely InfoZie, DataReduction and NBPredictor which process together to predict the recommendation list of expert developer for fixing the bug. This framework focus on domain specific environment in order to extend the idea for practical use. This system improves the accuracy of bug triaging and also, the quality of development and maintenance process in software industries.