<![CDATA[This study aims to analyze and compare the performance of the Dempster-Shafer and Certainty Factor methods in expert systems for diagnosing diabetes. The research uses a qualitative approach with a literature study method by reviewing various scientific publications related to both methods in medical expert systems. The analysis focuses on key aspects such as diagnostic accuracy, ability to handle uncertainty, computational complexity, and ease of implementation. The results show that the Certainty Factor method is more efficient and easier to implement, making it suitable for structured data with lower uncertainty, while the Dempster-Shafer method is more effective in handling complex uncertainty and incomplete data due to its evidence-based approach, although it requires more complex computations. The study concludes that no single method is universally superior, as each method has its own strengths depending on data characteristics and system requirements, and suggests that combining methods could improve the performance of expert systems in diabetes diagnosis.]]>
Copyrights © 2026
