The Users Emotional Study of Netease Cloud Music Based on LDA Model

Music sharing and social functions span the boundaries of language and cultural communication between people, and through the development of the Internet, the choice of music has been upgraded from the traditional record player mode to the discovery music mode of the music social platform. Based on the emotional needs users in “music social”, this study selects Netease cloud music reviews as data set. Through the optimized LDA model, Word2vec word vector model and sentiment dictionary, the topics are quantified from the perspectives of semantics and emotions respectively. In this way, we can explore the distribution characteristics of users' emotions under various music. At the same time, the music platform can conduct targeted emotional marketing and personalized recommendation through the emotional needs of users, which can promote user retention and improve marketing efficiency.