中文摘要：

提出一个新的链路预测相似性指标——局部社团结构指标（Local Community Structure,LCS）.在已知网络局部信息的前提下,LCS指标刻画了网络中任意两个节点的共同邻居节点与这两个节点的聚集关系.在基于真实网络的实验中,我们计算并比较了CN、AA、RA和LCS四个指标在7个不同真实网络中的AUC评价指标,发现在簇系数较大的真实网络中LCS指标的预测结果好于其他三个指标.

英文摘要：

We presented a new similarity measurement,the local community structure（LCS）in networks.Under the premise of local information on networks,the LCS measurement characterized the relationship between any two nodes in a network with their common neighbors.By a comparison of AUC,CN,AA,RA and LCS in seven real networks,we found that the accuracy of LCS was better than the other three measurements in the networks with big clustering coefficients.