<![CDATA[Stroke is a leading cause of disability and mortality worldwide, necessitating immediate and accurate first aid to mitigate severe outcomes. In Indonesia, limited public knowledge about stroke management, particularly among high school students, underscores the urgent need for targeted educational interventions. This study aims to evaluate students’ understanding of stroke first aid and identify optimal methods for clustering educational data using Agglomerative Hierarchical Clustering (AHC). A validated questionnaire was distributed to 112 high school students, focusing on their knowledge of stroke symptoms, risk factors, and first-aid practices. Data preprocessing ensured quality and consistency before applying AHC with three linkage methods: Single Linkage, Complete Linkage, and Ward’s method. The results were evaluated using Davies-Bouldin Index and Silhouette Coefficient to determine the most effective clustering approach. Ward’s method outperformed other linkage methods, achieving superior cluster compactness and separation. Four clusters were identified, representing varying levels of knowledge, from basic understanding to high awareness of stroke and seizure management. These findings provide a foundation for designing tailored educational programs, addressing specific knowledge gaps, and enhancing firstaid preparedness. This study demonstrates the utility of machine learning in educational research and contributes to improving public health education. Future research should expand on these findings by incorporating diverse datasets and alternative clustering algorithms.]]>
Copyrights © 2025
