Multi-Criteria Decision-Making GRA Approach for Selecting Appropriate Classification Algorithm in Software Defect Prediction Systems

Grey Relational Analysis (GRA) is a powerful Multi-Criteria Decision-Making (MCDM) method used to evaluate and rank classification algorithms based on their performance in software defect prediction. This study applies GRA to multiple datasets, including AR1, JM1, and CM1, to assess the effectiveness of different classifiers in identifying software defects. The analysis considers various performance criteria, such as accuracy, computational efficiency, and scalability, to determine the most suitable classification algorithm. The results indicate that classifiers like Lazy-IBK and Misc-HyperPipes consistently achieve higher GRA scores, demonstrating their superior predictive capability. The findings validate the significance of GRA in selecting the optimal classification model, highlighting its role in enhancing decision-making in software defect prediction.