<![CDATA[Neurological disorders present diagnostic challenges because of overlapping symptoms and limited access to specialist services in some healthcare settings. This study develops and evaluates a web-based expert system for supporting the preliminary diagnosis of five conditions—Stroke, Vertigo, Migraine, Low Back Pain, and Arthritis—using the Certainty Factor (CF) method. The CF approach incorporates Measure of Belief (MB) and Measure of Disbelief (MD) to represent uncertainty in the relationship between symptoms and diseases. The study adopts a design and development research (DDR) approach, with system development supported by an expert system development life cycle. The diagnostic performance was evaluated using 12 test cases, with the system results compared with assessments from an expert neurologist. The system correctly matched the expert assessment in 11 of 12 cases, resulting in a diagnostic agreement rate of 91.67%. One mismatch occurred in Case 2, in which the system identified Vertigo while the expert assessment indicated early-stage Stroke, reflecting the difficulty of distinguishing conditions with overlapping symptoms using the defined rules. Black-box testing showed that all tested functional scenarios passed, resulting in a 100% functional test pass rate. The developed system can support preliminary neurological disease identification by providing diagnosis results accompanied by Certainty Factor values.]]>
