Construction and Empirical Study of Music Education Model Based Upon Big Data and Relevance Theory

In the era of big data, music education faces the challenge that traditional teaching modes cannot meet students' individual needs. This study aims to integrate big data technology with Connectivism learning theory to construct and validate a new “big data + music education” teaching model. K-means clustering, linear regression, and Apriori association rule mining are applied to systematically evaluate the effectiveness of the model. The study develops a student-centered, data-driven, and intelligently matched teaching ecosystem; designs a whole-process instructional framework covering pre-class, in-class, and post-class stages; and establishes a multidimensional dynamic evaluation mechanism based on process data. This study verifies the practical value of big data technology in improving instructional precision and personalization in music education while providing a replicable and scalable theoretical framework and practical reference for art education reform in the context of educational informatization.