Intelligent Content Generation Mechanism for Diversified Music Teaching and Learning

Abstract

Diversified teaching is an inevitable requirement of music education reform, and increasing the diversity of teaching content generation is one of the ways to realize it. To this end, this paper proposes a DRMF diversification recommendation algorithm, which regulates the distribution of students’ ratings for different educational resource categories by optimizing the regularization parameter terms so as to realize the diversification of recommendation results. On this basis, a W2V-MMR teaching content generation algorithm is proposed, which utilizes Word2Vec model generation to train the word embedding model with big data text and then the word vectorization learning, which strengthens the semantic connection between the words so as to improve the accuracy of the teaching content generation. It is found that the ILD values of the DRMF algorithm are all higher than 0.215, the RMSDE values are all lower than 0.02, and the AD values are all around 1300. The W2V-MMR algorithm achieves 83.02% of teachers’ satisfaction after multiple rounds of optimization searching, and none of the MDEE values exceeds 0.4. It can be seen that both algorithms proposed in this paper yield good results.