Increasing Student Interest in Learning through the Implementation of the K-Nearest Neighbor Algorithm in Classifying Learning Preferences at SMAN 1 Kraksaan

Abstract

This research examines the effectiveness of implementing the K-Nearest Neighbor (KNN) algorithm in classifying student learning preferences and its impact on increasing interest in learning at SMAN 1 Kraksaan. The main aim of the research is to optimize learning methods through personalization based on individual student preferences. The study involved 560 students of SMAN 1 Kraksaan, with data including variables of age, gender, academic grades, daily study time, attendance and participation in class. The KNN algorithm is used to classify students' learning preferences into visual, auditory, kinesthetic, and reading/writing categories. The learning method is then adjusted based on the results of this classification. The results show that the KNN algorithm is able to classify student learning preferences with an accuracy of 80.36%. After implementing personalized learning methods, there was a significant increase in students' interest in learning, with an average increase of 1.76 points on a 10-point scale. Paired t-test analysis showed a statistically significant difference between interest in learning before and after intervention (p < 0.0001). This research concludes that the implementation of the KNN algorithm in classifying learning preferences can help increase students' interest in learning effectively. These findings emphasize the importance of personalization in education and demonstrate the potential of integrating machine learning in the pedagogical process to improve learning outcomes.