Emotional Behavior Analysis of Music Course Evaluation Based on Online Comment Mining

Abstract

This study investigates the method of analyzing emotional tendencies in music courses and its application in lesson plan evaluation. Using a weighted method to analyze emotional tendencies in music curriculum, the study compares the results with existing literature, demonstrating the superior accuracy of the proposed method. To evaluate lesson plan quality, a combination of self-assessment, mutual evaluation, group evaluation, and the middle school music lesson plan evaluation form is recommended for comprehensive assessment. The study's method for comment polarity achieves an accuracy rate of 69.19%, significantly outperforming other methods. Additionally, improvements in lexical feature extraction reduce computation complexity and interference factors in sentiment polarity analysis. In conclusion, this study offers valuable insights for enhancing teaching effectiveness, lesson plan quality, and understanding course feedback.