Automatic bug triage using text categorization

Davor Cubranic, Gail C. Murphy

Proceedings of the Sixteenth International Conference on Software Engineering and Knowledge Engineering (SEKE'04), June 2004, to appear.

Abstract

Bug triage, deciding what to do with an incoming bug report, is taking up increasing amount of developer resources in large open-source projects. In this paper, we propose to apply machine learning techniques to assist in bug triage by using text categorization to predict the developer that should work on the bug based on the bug's description. We demonstrate our approach on a collection of 15,859 bug reports from a large open-source project. Our evaluation shows that our prototype, using supervised Bayesian learning, can correctly predict 30% of the report assignments to developers.

KSI Copyright Notice: Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.