中文摘要：

语义社会网络是一种由信息节点及社会关系构成的新型复杂网络,传统语义社会网络分析算法在进行社区挖掘时需要预先设定社区个数,且无法发现重叠社区.针对这一问题,提出一种面向语义社区发现的link-block算法.该算法首先以LDA模型为语义信息模型,创新性地建立了以link为核心的block区域LBT（link-block-topic）取样模型;其次,根据link-block语义分析结果,建立可度量link-block区域的语义链接权重方法,实现了语义信息的可度量化;最后,根据语义链接权重建立了以link-block为单位的聚类算法以及可评价语义社区的SQ模型,并通过实验分析,验证了该算法及SQ模型的有效性及可行性.

英文摘要：

Since the semantic social network （SSN） is a new kind of complex networks, the traditional community detection algorithms which require presetting the number of the communities, cannot detect the overlapping communities. To solve this problem, an overlapping community structure detecting algorithm in semantic social networks based on the link-block is proposed. First, the measurement of the semantic weight of links for the link-block is established depending on the analysis of LBT. Secondly, a method to measure the semantic links weight of link-block area is developed to provide the measurement of semantic information. Thirdly, the overlapping community detection cluster method is designed, based on the semantic weight of links, with the link-block as the element. Finally, the SQ modularity for the measurement of semantic communities is obtained. The efficiency and feasibility of the algorithm and the semantic modularity are verified by experimental analysis.