中文摘要：

为对复杂网络进行合理划分,找出真实存在的社团结构,提出一种基于局部模块度和相似度的社团划分算法。计算网络中相连节点之间的相似度,快速聚合关联性最高的节点,从而实现社团的初步划分。以局部模块度为阈值,根据社团相似度聚合社团,得到具有最佳模块度的结果,避免模块度缺陷,提高算法准确度。算法进行社团划分时只需要网络局部信息,降低了时间复杂度。在实际网络和计算机仿真网络实验中的应用结果表明,与NS1,CNM,LAP等算法相比,该算法具有较低的计算复杂度和较高的准确率。

英文摘要：

The partition of complex network is useful to find the real community structure and help people research the real world, so this paper proposes a new algorithm of communities partition in network based on the local module degree and similarities. The algorithm calculates the similarities between the nodes in the network and aggregates the nodes which have the highest correlation. It aggregates the communities based on the local module degree and community similarity. Then the results is gotten which has the best local module degree. To verify the validity of the algorithm, this paper applies it in the real network and computer simulation network, and compares the algorithm with other algorithms. The result shows that the proposed algorithm has lower computational complexity and higher accuracy compared with NS1, CNM, LAP algorithms, etc..