Comparative Study of KNN and SVM Methods for Analyzing College Major Consistency Based on High School Background

Selecting a college major that aligns with students’ high school background is an essential factor in supporting academic achievement and career preparation. This study focuses on a comparative analysis of the Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) algorithm ms in evaluating the consistency of college major selection. A dataset of 636 students was collected and processed for analysis. Model evaluation was performed using 5-Fold Cross Validation, in which the dataset was repeatedly partitioned into training and testing sets to ensure reliable and unbiased performance assessment. The results suggest that SVM demonstrates higher effectiveness, achieving average scores across precision, recall, F1-score, and accuracy of 85%. Meanwhile, KNN obtained average performance scores of 78%. These findings highlight that SVM provides better performance in analyzing the consistency between students’ high school majors and their chosen college majors. These findings also contribute to the development of decision support systems and counseling services to guide students in making more informed major choices.