Analisis Tipe Kecerdasan Majemuk Siswa Sekolah Dasar Berbasis Catatan Perilaku Menggunakan Algoritma Naive Bayes, K-Nearest Neighbors, dan Support Vector Machine

Abstract

This study aims to identify the types of multiple intelligences of elementary school students based on Howard Gardner's theory by utilizing machine learning algorithms, namely Naive Bayes, K-Nearest Neighbors (KNN), and Support Vector Machine (SVM). The data used comes from student behavior records and intelligence type questionnaires obtained from students or parents. The SEMMA method (Sample, Explore, Modify, Model, Assess) is used, including text preprocessing and TF-IDF feature extraction. The classification process is carried out using Orange Data Mining software and evaluated using accuracy, precision, recall, F1-score, and AUC metrics. The evaluation results show that the SVM algorithm provides the best performance with an accuracy of 93.30% and AUC of 0.997. Naive Bayes follows with 90.50% accuracy and 0.994 AUC, while KNN reaches 89.50% accuracy and 0.941 AUC. The study also results in a web-based application prototype that classifies students' intelligence types and provides personalized learning recommendations. This confirms the effectiveness of machine learning in supporting personalized learning and student potential development.