Bug fix-time prediction model using naïve Bayes classifier

Predicting bug fix-time is an important issue in order to assess the software quality or to estimate the time and effort needed during the bug triaging. Previous work has proposed several bug fix-time prediction models that had taken into consideration various bug report attributes (e.g. severity, number of developers, dependencies) in order to know which bug to fix first and how long it will take to fix it. Our aim is to distinguish the very fast and the very slow bugs in order to prioritize which bugs to start with and which to exclude at the mean time respectively. We used the data of four systems taken from three large open source projects Mozilla, Eclipse, Gnome. We used naïve Bayes classifier to compute our prediction model.