<![CDATA[Student academic achievement is an important indicator for assessing the success of the learning process in higher education. Students’ exam scores are influenced not only by academic factors, such as study hours, class attendance, study methods, exam difficulty, and course, but also by non-academic factors, such as age, gender, internet access, sleep duration, sleep quality, and learning facilities. This study aims to compare the performance of the K-Nearest Neighbor (KNN) and Naïve Bayes algorithms in classifying students’ exam scores into two categories: Poor and Good. The dataset used in this study consists of 5,716 records with 12 attributes. The research process includes data preprocessing, target transformation, data splitting into training and testing sets, algorithm implementation, and performance evaluation using accuracy, precision, recall, and classification error. The results show that Naïve Bayes achieved an accuracy of 83.02%, precision of 82.85%, recall of 82.79%, and classification error of 16.98%, making it more effective than KNN in classifying students’ exam scores.]]>
Copyrights © 2026
