Innovative machine learning integration for sustainable software quality and performance optimization in the SDLC

This innovative research contributes to the sustainability of software development by addressing critical challenges in software defect analysis, performance evaluation, and quality prediction within the Software Development Life Cycle (SDLC). By integrating cutting-edge machine learning techniques into quality assurance processes, we present a novel approach that enhances the long-term viability and reliability of software projects. We developed predictive models using supervised machine learning techniques, including Random Forest and Random Forest Regressor. These models achieved an impressive defect prediction accuracy of 87.4% and demonstrated precision in performance prediction with a Mean Squared Error (MSE) of 0.128 and an R @@2 score of 0.84. By analyzing code metrics and leveraging machine learning, the study provides actionable insights for developers to improve resource allocation and software reliability. These findings underline the significance of integrating machine learning into software engineering practices, setting a foundation for future advancements in quality assurance methodologies.