A Data-Driven Framework for Music Aesthetic Education Content Optimization

Abstract

To address the problems of homogenized content, delayed feedback, and insufficient alignment between teaching content and student interests in music aesthetic education, this study presents a data-driven framework for curriculum content optimization. The framework integrates multi-source student data and applies K-means clustering to build student personas, while using a content-matching mechanism and sentiment analysis to support adaptive recommendations and feedback optimization. A four-week comparative teaching experiment involving 150 undergraduate students from a cooperating university demonstrates that the proposed framework can improve student satisfaction and engagement compared with traditional teaching approaches. The study also discusses its adaptability across different educational contexts and offers a practical approach to personalized, intelligent development of music aesthetic education.