Retrieval Algorithm Model of Music Preschool Teaching Resources Based on Cloud Platform Resources

In recent years, the development and application of cloud computing technology in many fields has made resource management more mature. In the field of education, due to the explosive growth of educational resources, it is also facing difficulties in resource retrieval, storage and sharing. At the same time, the wide application of mobile terminals makes the self-upgrade, synchronization and protection of personal resources and applications face great challenges. The cloud platform is based on the development of cloud computing and can be used to provide a network platform under the co-construction and sharing mode of various resources. The basic idea of the technology is to push the server to the edge network by adding a new network architecture on top of the existing Internet. For the music retrieval with the whole song content as the query condition, a model based on manifold sorting is proposed, and the retrieval results are improved by designing relevance feedback. Gaussian mixture model and maximum likelihood estimation are used to cluster the spectral data of each audio, and the center of each cluster is selected as the typical spectral feature. The application of cloud computing technology in the field of education is called “education cloud”. By virtualizing educational information resources and systems and uniformly deploying and implementing them on the cloud platform, cloud services can be provided for education practitioners and students.