Analisis Reaksi Emosional Siswa terhadap Pembelajaran SAVI Menggunakan Support Vector Machine

Abstract

The development of information and communication technology has enabled the implementation of more interactive and personalized learning methods, one of which is the SAVI (Somatic, Auditory, Visual, Intellectual) learning model. This study aims to analyze students' emotional responses to SAVI-based learning using the Support Vector Machine (SVM) algorithm for sentiment classification. The study involved 100 student respondents who provided opinions regarding their learning experiences. Prior to analysis, the opinion data underwent preprocessing, including case folding, cleaning, tokenization, stopword removal, and stemming, to ensure high-quality features for the SVM model. The classification results indicate that the majority of students (65%) showed positive sentiment, while 20% expressed negative sentiment and 15% neutral. Integration of the SAVI model revealed that students with Somatic learning styles tended to show positive sentiment, whereas students with Visual and Intellectual learning styles exhibited a more varied sentiment, including negative and neutral. The SVM model performance evaluation demonstrated high precision, recall, and F1-score, particularly for the majority class, indicating the model's accuracy in classifying student opinions. These findings highlight the importance of considering students' emotional responses in SAVI-based learning, as emotional factors significantly influence motivation and learning outcomes. The integration of SVM and SAVI provides comprehensive insights for designing adaptive, responsive, and data-driven learning strategies to enhance learning effectiveness.