<![CDATA[In Indonesia, school accreditation is a crucial process for assessing the eligibility of educational institutions to meet national education standards. However, this process is resource-intensive and requires significant time, manpower, and financial resources. This study aimed to explore the application of machine learning classification methods: Random Forest, Boosting, and Support Vector Machine (SVM) to predict the accreditation ranks of Junior High Schools in Indonesia. The goal was to create an efficient, automated model to predict school accreditation status, improve the efficiency of the accreditation process, and facilitate better resource allocation. Data preparation included handling missing values, reducing the data dimensions, and addressing data imbalances. The dataset consisted of 23,954 Junior Schools from 34 provinces, with 37 variables, including 36 predictors and one target variable (accreditation status). The study found that Random Forest outperformed Boosting and SVM, with the highest Area Under Curve (AUC) of 0.8133. Random Forest also demonstrated the lowest average classification error of 19.32%, indicating its superior performance in predicting junior high school accreditation ranks. The results suggest that machine learning models, particularly Random Forest, can provide a more efficient and reliable alternative to manual accreditation evaluations. This approach can optimize educational assessments, improve resource allocation, and offer valuable insights for policymakers to enhance school performance, particularly in under-served regions.]]>
Copyrights © 2026
