Fuzzy Soft Set-Based Risk Classification of Gastric and Prostate Cancer Patients Using Mamdani and Sugeno Models

This paper presents a hybrid decision-support framework for the risk classification of gastric and prostate cancer patients using fuzzy soft set theory combined with Mamdani and Sugeno fuzzy inference models. The inherent uncertainty in medical data, such as imprecise symptoms and overlapping diagnostic indicators, is effectively addressed through fuzzy soft sets, which integrate the flexibility of soft set theory with the vagueness-handling capacity of fuzzy logic. Clinical parameters were transformed into fuzzy linguistic variables and structured into fuzzy soft representations. These were then evaluated using rule-based Mamdani and function-based Sugeno inference systems to categorize patients into low, moderate, and high-risk groups. Experimental results demonstrate that the Sugeno model achieves higher numerical accuracy and computational efficiency, while the Mamdani model offers superior interpretability. The proposed approach achieves robust classification performance across both cancer types and holds potential for integration into intelligent clinical decision support systems. This study contributes a novel comparative evaluation of fuzzy inference models within a fuzzy soft set framework, addressing a critical challenge in uncertainty-aware medical diagnostics.