Published Paper Details:

Software bugs are a critical challenge in the development lifecycle, impacting product quality, user satisfaction, and maintenance costs. Traditional bug prediction models often focus on static code metrics or historical bug data, without considering the developer-centric context. This paper proposes a Developer-centered bug prediction model that leverages both code-level features and developer-specific behavioral patterns to enhance prediction accuracy. The model integrates features such as code complexity, change history, developer expertise, and commit frequency, using advanced machine learning techniques like ensemble learning and deep neural networks. We validate our approach on real-world datasets from open-source repositories, demonstrating significant improvements over conventional models in terms of precision, recall, and F1-score. The results suggest that incorporating developer-related factors provides a more nuanced understanding of bug-prone code, enabling more targeted and effective bug management strategies.