Convolutional Neural Network based Recommendation Algorithm of Instructional Resources Mechanism for Online Music Education

The forms of music curriculum resources are very diverse, and the broad Internet platform provides many valuable materials for the integration of music curriculum resources. As an important part of tertiary education informatization, online instructional resources play an increasingly important role in promoting learners and educators' knowledge construction, improving their practical ability and developing their advanced thinking ability. In this article, a recommendation algorithm of instructional resources in universities based on computer-aided technology and deep learning (DL) is proposed. Convolutional neural network (CNN) is integrated into the joint probability matrix decomposition model to fully mine the hidden information in existing instructional resources and instructional resources. The results show that after continuous training, the accuracy of proposed algorithm is significantly higher than that of the other two algorithms, reaching over 95%. Compared with information retrieval technology, the recommendation of instructional resources can better meet the individual needs of students. Therefore, it is feasible to apply the model to the recommendation of instructional resources mechanism, which can provide modern theoretical and technical support for the reform of music education mechanism. © 2023 CAD Solutions, LLC,.