Hybridizing Grid Partitioning and Rough Set Method for Fuzzy Rule Generation: A Robust Framework for Dataset Classification with Enhanced Interpretability and Scalability

Abstract

This research presents a novel approach, called GP-RS-FRG, that combines grid partitioning and rough set method for fuzzy rule generation in dataset classification. The aim is to enhance interpretability and scalability while maintaining accuracy in the classification process. Traditional classification methods often lack transparency, making it difficult to interpret their decisions, especially with complex datasets. Additionally, these methods may face challenges in handling large datasets with numerous attributes and instances. The proposed framework addresses these limitations by generating transparent and understandable fuzzy rules. The GP-RS-FRG framework utilizes grid partitioning to divide the input space into non-overlapping grid cells, reducing the search space and improving computational efficiency. By integrating the rough set method, the framework identifies the most significant attributes, reducing redundancy and simplifying the rule base. This enhances interpretability and simplifies the decision-making process. The generated fuzzy rules capture the complex relationships between attributes and classes, providing meaningful insights into the classification model. Experimental evaluation on diverse datasets demonstrates the effectiveness of the GP-RS-FRG framework in generating accurate fuzzy rules while maintaining interpretability and scalability. The framework enables domain experts to understand and interpret the classification process, facilitating informed decision-making. It has potential applications in various domains where transparent and scalable classification models are required. Future research directions may include exploring alternative approaches, variations, or refinements to further enhance the framework's performance. Comparative studies and experiments on larger and more diverse datasets would provide a deeper understanding of its capabilities and limitations. The generalizability and applicability of the framework to different domains should also be investigated to promote wider adoption and impact.