Sentiment Analysis of GitHub Commit Comments using Machine Learning Techniques

The proliferation of collaborative platforms like GitHub necessitates robust sentiment analysis tools to interpret developer emotions in commit comments, offering insights into project health and team dynamics. This study advances existing research by evaluating three machine learning algorithms for this task: Passive Aggressive Classifier, C-Support Vector Classifier (C-SVC) with a linear kernel, and Random Forest Classifier.We analyzed 10,000 GitHub commit comments from diverse projects spanning multiple programming languages and development periods. Performance was assessed using accuracy, precision, recall, and F1-score. C-SVC outperformed other algorithms with 98 percent accuracy, 97 percent precision, 98 percent recall, and an F1-score of 0.975. The Random Forest Classifier achieved 96 percent accuracy, while the Passive Aggressive Classifier reached 94 percent.Our findings demonstrate C-SVC’s effectiveness for sentiment analysis in this context, surpassing previous benchmarks. Integration of this tool into development workflows could enhance team communication, enable early issue detection, and provide quantitative metrics for project emotional health. Future research should examine the generalizability of these results across various project types and explore real-time sentiment analysis in collaborative coding environments.