Issue Type

Issue type classification focuses on automatically categorizing software development issues (e.g., bugs, feature requests, questions) reported in issue tracking systems. Current research emphasizes using machine learning, particularly transformer-based models like BERT and RoBERTa, to improve the accuracy and efficiency of this classification, often achieving high F1-scores. This automated categorization significantly reduces manual effort in software maintenance, leading to faster resolution times, cost savings, and improved software quality.