A comparative study of machine learning algorithm for software defect prediction

Key Points

• Improved accuracy results from hybrid models integrating various machine learning techniques, enhancing defect prediction.
• The hybrid machine learning model includes Random Forest, SVM, CNN, and Logistic Regression for superior performance.
• Analysis utilizes multiple machine learning algorithms to assess prediction accuracy for software defects, suggesting advancements over traditional methods.
• Results underline the importance of integrating diverse machine learning algorithms in software defect prediction approaches.