A REVIEW ON BUG REPORT CLASSFICATION USING MACHINE LEARNING APPROACHES

Abstract :

Bug reports are crucial documents in software development, providing detailed information about software issues, including descriptions, current status, and how severe they are. They are essential for identifying and keeping track of software problems, which are essential for ensuring the overall quality of software systems. When there are no unresolved bugs, it's a clear indicator that the software is reliable and works smoothly. In recent years, machine learning has becomeincreasingly skilled at classifying different types of software bugs. One way it does this is by using ensemble machine learning models that combine various tools like random forests, decision trees, and naive Bayes. Additionally, support vector machines (SVM), decision trees (C4.5), and other classification methods have been used to better understand and categorize bugs. This paper focuses on how machine learning can revolutionize bug tracking, nature of bug, contributing to the ongoing conversation about making software more dependable.