<![CDATA[Determining the educational trajectory following junior high school graduation represents a pivotal decision shaped by students' academic competence, personal interests, and inherent personality inclinations. In practice, this selection process is frequently carried out in a subjective manner, which risks producing a disconnect between students' genuine potential and their eventual educational placement. The present research seeks to examine and compare the predictive performance of the Artificial Neural Network (ANN) and Support Vector Machine (SVM) algorithms in forecasting students' preference for either senior high school (SMA) or vocational high school (SMK) among learners at SMP Panca Budi Medan. A total of 220 student records were employed as the dataset, incorporating academic performance data alongside RIASEC personality scores as the predictor variables. All data processing was executed within the WEKA application environment utilizing 10-fold cross validation as the evaluation scheme. The ANN model was constructed through the Multilayer Perceptron approach, while SVM relied on the Sequential Minimal Optimization (SMO) technique. Experimental findings revealed that both classifiers attained an identical accuracy rate of 85.45%; however, the ANN model demonstrated a superior ROC Area value of 0.925 relative to the SVM's 0.849, signifying that ANN possesses stronger discriminative capability in distinguishing SMA from SMK selections. The study confirms that integrating academic metrics with RIASEC scores provides a viable foundation for constructing a machine learning-driven school-choice prediction system that is both more objective and better attuned to individual student profiles.]]>
Copyrights © 2026
