<![CDATA[Refractive eye disorders remain one of the leading causes of visual impairment worldwide, while limited access to ophthalmology services often delays early diagnosis. This study investigates the effectiveness of the Decision Tree algorithm for symptom-based classification of refractive eye disorders and its implementation within a web-based expert system for preliminary eye health screening. Data were collected from 505 respondents using a structured Google Forms questionnaire. After preprocessing and labeling, the dataset was divided into 80% training data and 20% testing data. Model performance was evaluated using a confusion matrix together with accuracy, precision, recall, and F1-score metrics to provide a comprehensive assessment under an imbalanced class distribution. Experimental results showed that the proposed Decision Tree model achieved an overall accuracy of 89.11% and demonstrated satisfactory performance in identifying dominant diagnostic classes while maintaining transparent and interpretable decision rules. The developed web-based system provides symptom-based diagnosis, examination history management, and PDF report generation to support preliminary eye health assessment. These findings indicate that the proposed approach is suitable as an accessible decision-support tool for preliminary refractive eye disorder screening, particularly in areas with limited access to ophthalmology services.]]>
Copyrights © 2026
