Community structure is an important property of the complex networks. How to detect the communities is significant to understand the network structure and analyze the network properties. Many algorithms, such as K-L and GN, have been proposed to detect the community structure in complex networks. However, the communities detected by these algorithms are always not overlapping. According to daily experience, a node in the network may belong to several communities and a community should have many nodes and connections. Based on these principals and existing researches, an efficient algorithm for overlapping community detection in complex networks is proposed in this paper. The key strategy of the algorithm is to mine a node with the closest relations with the community and assign it to the community. Some real-world networks are used to test the performance of the algorithm. Experimental results demonstrate that the algorithm proposed is rather efficient to detect the overlapping community in complex networks.