<![CDATA[Students’ academic achievement is an important indicator of the success of the educational process; however, its assessment is often subjective and not yet fully data-driven. Therefore, a systematic analytical approach is required to classify students’ academic achievement objectively and accurately. This study aims to compare the performance of Support Vector Machine (SVM) and Naive Bayes algorithms in classifying the academic achievement of grade III students at SMAS BPD Tobelo Selatan. A data mining approach using classification techniques was applied, involving 17 attributes as predictor variables and two target classes of academic achievement, namely Very Good and Good. Data processing and model evaluation were conducted using the WEKA software, with performance measured through accuracy, precision, recall, and the Area Under the Receiver Operating Characteristic Curve (AUC-ROC). The results indicate that the SVM algorithm achieves the best performance in terms of accuracy, precision, and recall, each reaching 97.78%, while the Naive Bayes algorithm obtains the highest AUC-ROC value of 98.08%. These findings demonstrate that SVM is superior in prediction accuracy, whereas Naive Bayes shows excellent capability in class discrimination. This study is expected to support data-driven academic decision-making in school environments.]]>
Copyrights © 2026
