Improving Fuzzy Rule Classifier with Brain Storm Optimization and Rule Modification

TL;DR

This research specifically addresses the challenge of diabetic classification by employing the Brain Storm Optimization (BSO) algorithm to propose a novel fuzzy system that redefines rule generation for this context, tailored specifically for diabetes-related data.

Abstract

The expanding complexity and dimensionality in the search space can adversely affect inductive learning in fuzzy rule classifiers, thus impacting the scalability and accuracy of fuzzy systems. This research specifically addresses the challenge of diabetic classification by employing the Brain Storm Optimization (BSO) algorithm to propose a novel fuzzy system that redefines rule generation for this context. An exponential model is integrated into the standard BSO algorithm to enhance rule derivation, tailored specifically for diabetes-related data. The innovative fuzzy system is then applied to classification tasks involving diabetic datasets, demonstrating a substantial improvement in classification accuracy, as evidenced by our experiments.

Figures (5)

• Figure 1: Example of membership functions.
• Figure 2: Depiction of Strategy for Addressing Challenges.
• Figure 3: Improved brain storm optimization algorithm flow chart.
• Figure 4: The impact of e and K on the classification performance of the system for PID.
• Figure 5: Sample membership function.