中文摘要：

为充分利用社交网络资源，以MetaFiher为数据源，分析社会语义网络结构。通过面向对象方法，指出社会语义网络是融合多维语义关系的元网络；以posttitle为语料，揭示tag分布符合幂律，并参照CiteULike，探讨用户标注行为及主题的演化；统计主流版块的comment、favorite及category数据，透视用户标注主题及方式；围绕整体网络、多维语义关系网络，借助相应指标量化网络结构特征和用户角色。用户角色识别不能单纯依靠网络分析，尚需结合语义分析。Q＆A是社会语义网络中最活跃的语义关系，但知识复用率较低，且基于实际问答关系的用户连通性不足。

英文摘要：

In order to take full advantage of social network resources, this paper takes MetaFilter as data source to analyze the structure of social semantic network. Through the object-oriented approach, the paper points out that the semantic social network is a meta-network, which integrates the multi-dimensional semantic relationship. The paper takes ＂post title＂ as corpus to reveal that the distribution of tag is in accordance with the power law. The paper discusses the evolution of user tagging behavior and topic by reference to the CiteULike. The paper counts the data of comment, favorite and category of the mainstream, and discusses the topic and methods of users tagging. Based on the overall network and multi-dimensional relation network, the paper quantifies the features of network structure and user role by using the corresponding indexes. The identification of user role cannot only depend on network analysis, but also need to combine the semantic analysis. Q＆A is illustrated to be the hottest semantic relationship in semantic so- cial network, but its knowledge reuse rate is low and user connectivity based on the actual Q＆A relations is insufficient.