本文将社会化标注系统分层分析,首层分成若干个社区,次层将主题作为主导因素,同时结合LDA(Latent Dirichlet Allocation)思想构造动态贝叶斯模型并将时间因素加入进来得到时间动态演变下的隐含社区及主题的标签集,对于为网络组织及用户有效获取提供信息资源的内在属性,提炼出社会化标注系统的有效信息。 In this paper, the social annotation system is hierarchically analyzed. The first floor is divided into several communities. The second floor uses the theme as the dominant factor. In the meantime, the dynamic Bayesian model is constructed by incorporating the LDA (Latent Dirichlet Allocation) and the time factor is added to obtain the dynamic evolution of the time Under the implicit community and the theme of the tag set, for the network organizations and users effective access to information resources to provide the intrinsic attributes of social labeling system to extract valid information.