INTELLIGENT BUG TRIAGING FOR OPEN-SOURCE SOFTWARE THROUGH MACHINE LEARNING TECHNIQUES

Abstract

In large-scale software development, bug triaging the process of categorizing and allocating software issues to the right developers is an essential but difficult activity. Manual triaging takes a lot of time, is inconsistent, and is subject to human bias. As a result, it frequently causes delays in issue resolution and misallocates developer resources. This study investigates the use of machine learning to automate and enhance the accuracy and efficiency of bug triaging. We assess a number of machine learning models, such as Multinomial Naive Bayes and Random Forest. We demonstrate a robust end-toend pipeline for data preprocessing, augmentation, model training, and evaluation using multi-label classification techniques, along with high prediction performance.