Innovative Path of Music Education Teaching in Colleges and Universities under the Architecture of Disciplinary Knowledge Mapping

Abstract

The purpose of this paper is to present a pedagogical model for mapping subject knowledge in the field of music education. The model includes two aspects: the extraction of entities and inter-entity relationships within the subject domain and the construction of a subject ontology. In the construction process of subject knowledge mapping, this paper proposes a remote supervised music subject knowledge extraction method based on a convolutional neural network combined with an attention mechanism, which realizes the extraction of entities in the music subject domain and the relations between corresponding entities. This paper proposes a keyword extraction method for obtaining the set of discipline concepts in ontology construction. To improve performance, this paper proposes a framework structure for knowledge fusion that includes three aspects: data preprocessing, similarity calculation, and knowledge fusion. As a result of the study, the teaching model can improve the performance of music majors on average by 13.3. In terms of self-efficacy, on average reached 3.58 (SD = 0.535), which is at a good level. The results demonstrated the effectiveness of the music teaching model based on subject knowledge mapping.