OPTIMIZING EXPERT SYSTEMS USING GENETIC ALGORITHMS: A PRACTICAL AI-BASED APPROACH

Abstract

Expert systems are computer programs that mimic human reasoning to make decisions in complex domains. These systems rely heavily on a predefined set of rules that can be difficult to optimize manually. Genetic Algorithms (GAs), inspired by natural selection, have shown great promise in solving complex optimization problems. This study explores the integration of GAs into expert systems to automatically evolve and improve their rule base. A prototype expert system for medical diagnosis was developed, and GA was used to optimize rule weighting and selection. Results demonstrated significant improvement in decision accuracy and system robustness. The findings suggest that GAs can enhance expert systems’ adaptability and performance, providing a strong basis for hybrid AI systems.