中文摘要：

在静态网络图中对社会网络进行分析，可能忽略网络的时间特性，从而错过捕捉动态网络演变模式的机会。为检测社区的演变及社区结构随时间的变化，分析动态社区，为每一个社区定义一系列重大事件，给出社区匹配算法，并对元社区的概念进行定义。实验结果证明，采用该算法建模和检测社会网络中的社区演变，可有效识别和追踪随时间变化的相似社区。

英文摘要：

To analyze social network in static network may omit the characteristic of time, so it fails to capture the evolutionary pattems in dynamic networks. In order to detect the community evolutions and the community structures that change in time, this paper analyzes the dynamic community, and a series of significant events is defined for each community. A community matching algorithm is proposed and it defines the concept of meta community. Experimental results show that applies the method to model and detect community evolution in social networks, can efficiently identify and track similar communities over time.