<![CDATA[This study applies the k-medoids clustering machine learning approach to assess regional clustering in Indonesia based on educational quality. Data on the quality of education, including indicators of school enrollment rate (APS), gross enrollment rate (APK), and pure participation rate (APM), is gathered and processed from all provinces in Indonesia. The k-medoids clustering technique is used to carry out the clustering process, while metrics like Dunn's index, connection coefficient, and silhouette score are used to evaluate the results. The study's findings indicate that three clusters are the ideal amount, with a silhouette score of 0.2388, a connectivity coefficient of 7.1405, and a Dunn's index value of 0.1651. Cluster homogeneity is likewise moderate, despite the regions' moderate distances from one another. This assessment offers a thorough understanding of Indonesia's educational quality clustering pattern, which can serve as a foundation for developing education strategies in different areas]]>
Copyrights © 2024
