Music Recommendation System Design Based on Gaussian Mixture Model
暂无分享,去创建一个
The paper establishes a double-layer classifier based on Gaussian Mixture Model and the Thayer model to divide the music style into several categories. On the basis of effective verification of experiments, the music listening experience is added into the model to analyze and normalize two-dimensional data points and the tastes of music users or playing times also can be added to obtain new three-dimensional data points. Then the Gaussian Mixture Model is employed again for classifying the new three-dimensional data points. In this way, not only can the taste changing process of users for different music be analyzed, but also the similarity among different users can be calculated. Therefore, music can be recommended properly to music users.
[1] Jaakko Astola,et al. The Mel-Frequency Cepstral Coefficients in the Context of Singer Identification , 2005, ISMIR.
[2] Markus Werner,et al. ATM-Based Routing in LEO/MEO Satellite Networks with Intersatellite Links , 1997, IEEE J. Sel. Areas Commun..
[3] Douglas A. Reynolds,et al. Speaker identification and verification using Gaussian mixture speaker models , 1995, Speech Commun..