基于用户标签的社区发现方法研究

摘要

In order to dig mining potential users' information and provide better personalized services, this paper proposes a user oriented community discovery algorithm based on analyzing users' usage of tags. We considered the frequency and sparsity of tags, create an adjacency matrix to cluster users, and complete community discovery at last. What' s more, through an experiment we demonstrate this algorithm is feasible.%为充分挖掘标签中潜在的用户信息、提供更好的个性化服务,在分析用户标签使用情况的基础上,提出一种面向用户的社区发现方法.综合考虑标签使用频率和稀疏度,并同时使用二者构造一个邻接矩阵进行聚类,最终完成社区发现,并且论证了该方法的可行性.