<![CDATA[Diarrheal disease in children remains a major public health problem in Indonesia, particularly among children under five years old. Limited access to healthcare services and medical experts in several regions often causes delays in the diagnosis and treatment process, increasing the risk of dehydration and mortality. Previous studies on expert systems in the healthcare field have generally focused on common diseases and have rarely discussed web-based diagnosis systems specifically designed for childhood diarrheal diseases using the Naïve Bayes method and real medical record data. Therefore, this study aims to develop a web-based expert system capable of assisting the early diagnosis of diarrheal diseases in children using the Naïve Bayes classification method. The research process involved literature studies, expert interviews, medical record data collection from 50 patients at Depok III Public Health Center, system design using Unified Modeling Language (UML), implementation using PHP and MySQL, and system validation testing. The Naïve Bayes method was applied to calculate the probability of seven categories of diarrheal diseases based on user-selected symptoms. The testing results demonstrated that the developed system achieved an accuracy level of 94% compared to expert diagnoses. These findings indicate that the proposed expert system can provide fast and accurate preliminary diagnoses and can support parents and healthcare workers in identifying diarrheal diseases in children more effectively.]]>
