Applying TOPSIS Method for Optimal Data Mining Algorithm Selection in Software Defect Prediction Systems

In software defect prediction systems, choosing the best data mining classification algorithm is essential for efficient decision-making. The Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method for ranking classification algorithms according to several criteria is assessed in this study. Classifiers were evaluated and ranked using the AR1 and JM1 datasets using a structured multi-criteria decision-making (MCDM) methodology. According to the results, Lazy-IBK continuously performs better than other classifiers, demonstrating how well TOPSIS finds the top-performing models. Furthermore, the comparative study of classification algorithms shows how reliable the approach is when taking into account variables like scalability, accuracy, and computational efficiency. According to the results, TOPSIS considerably enhances classifier selection when it comes to software defect prediction, resulting in data mining applications that make better-informed and impartial decisions. Key Words: MCDM, Software Defect Prediction, TOPSIS