Hybrid Grid Partition and Rough Set Method for Fuzzy Rule Generation in Dataset Classification: Enhancing Accuracy and Interpretability

Abstract

This research presents a hybrid grid partition and rough set method for fuzzy rule generation in dataset classification, aiming to enhance accuracy and interpretability. The proposed mathematical model combines grid partitioning, rough set theory, and fuzzy logic to identify relevant attributes, reduce dimensionality, and generate interpretable fuzzy rules. The model is evaluated using a case example of iris flower classification and demonstrates competitive accuracy in predicting the species of iris flowers based on their attributes. The interpretability of the generated fuzzy rules provides transparent explanations for the classification decisions, allowing domain experts to understand and interpret the reasoning behind the predictions. Comparative analysis with traditional algorithms showcases the superiority of the hybrid model in terms of accuracy and interpretability. Sensitivity analysis enables parameter tuning and customization, further improving the model's performance. The practical implications of the hybrid model are discussed, and its potential applications in various domains are highlighted. The research concludes that the hybrid grid partition and rough set method offer an effective approach for accurate and interpretable dataset classification, with implications for decision-making and insights in real-world applications.