Intelligent Bug Triaging in Open-Source Software Using Machine Learning Techniques

Authors

Keywords:

Machine Learning, Bug Triaging, Open-Source Software, Natural Language Processing, Software Maintenance, Bug Classification, Predictive Analytics

Abstract

This paper examines the potential of machine learning to improve the accuracy and efficacy of bug assignment. In particular, it underscores the importance of meticulous defect triaging in open-source software. On a daily basis, thousands of problem reports are submitted to extensive open-source software repositories. Manual classification is time-consuming, error-prone, and inefficient. The objective of this investigation is to assess the efficacy of machine learning techniques, including Naïve Bayes, SVM, Random Forest, and deep learning models, in the identification of problem reports and their assignment to developers. This classification employs project activity, developer expertise, textual descriptions, and issue history. By extracting valuable information from problem reports, Natural Language Processing (NLP) assists in the anticipation and prioritization of work. Our objective is to establish an automated and scalable bug remediation framework that will enhance the quality of software, reduce the time required to resolve defects, and increase the productivity of developers. Experiments conducted with authentic open-source datasets demonstrate that bug triaging based on machine learning enhances assignment precision and minimizes manual labor.

Author Biography

  • Dr. KISHOR KUMAR GAJULA, MOTHER THERESSA COLLEGE OF ENGINEERING & TECHNOLOGY

    Associate Professor, Department of CSE,

    MOTHER THERESSA COLLEGE OF ENGINEERING & TECHNOLOGY, PEDDAPALLI,

     

Downloads

Published

2026-11-11