It's not a bug, it's a feature: how misclassification impacts bug prediction

Abstract

This submission presents work submitted and accepted at the International onference on Software Engineering in 2013 [Hj2013]. In empirical software engineering, it has become common to mine historic data to detect where bugs have occurred in the past, or to predict where they will occur in the future. The accuracy of such models depends on the quality of the data. For example, defect prediction models rely on the accuracy of historic data, such as bug reports.