<![CDATA[This study develops an interpretable rule-based clinical decision support system for early screening of heart disease presence by integrating the C4.5 decision tree algorithm with a rule-based reasoning mechanism. The proposed approach is intended to assist clinicians in obtaining rapid, transparent preliminary indications from clinical data, particularly in settings that require lightweight and auditable solutions. The dataset was obtained from the UCI Heart Disease Repository and comprises 299 patient records, evaluated using a 70% training and 30% testing split in RapidMiner. Experimental results show that the C4.5 model achieves an accuracy of 86.52% and produces clinically interpretable IF–THEN rules, enabling traceable reasoning and decision auditing. Although C4.5 is a classical learning algorithm, it remains relevant for clinical decision support due to its auditability, low computational cost, and ease of deployment in resource-constrained environments. The developed system is expected to support early screening/triage and data-driven clinical decision-making, contributing to the advancement of medical decision support systems (MDSS).]]>
Copyrights © 2026
