Rule-Based Modeling of Low-Dimensional Data with PCA and Binary Particle Swarm Optimization (BPSO) in ANFIS

• DOI: 10.48550/arxiv.2502.03895
• Publication Date: 2025-02-06
• Authors: Afnan Al-Ali; Uvais Qidwai
• Keywords: Particle (ecology)

Fuzzy rule-based systems interpret data in low-dimensional domains, providing transparency and interpretability. In contrast, deep learning excels complex tasks like image speech recognition but is prone to overfitting sparse, unstructured, or data. This interpretability crucial fields healthcare finance. Traditional systems, especially ANFIS with grid partitioning, suffer from exponential rule growth as dimensionality increases. We propose a strategic rule-reduction model that applies Principal Component Analysis (PCA) on normalized firing strengths obtain linearly uncorrelated components. Binary Particle Swarm Optimization (BPSO) selectively refines these components, significantly reducing the number of rules while preserving precision decision-making. A custom parameter update mechanism fine-tunes specific layers by dynamically adjusting BPSO parameters, avoiding local minima. validated our approach standard UCI respiratory, keel classification, regression datasets, real-world ischemic stroke dataset, demonstrating adaptability practicality. Results indicate fewer rules, shorter training, high accuracy, underscoring methods effectiveness for scenarios. synergy fuzzy logic optimization fosters robust solutions. Our method contributes powerful framework interpretable AI multiple domains. It addresses dimensionality, ensuring base.