Abstract: The scientific research cooperation is becoming more and more popular. There are a large number of complex networks, such as citation networks, co-citation networks and co-author networks and so on. Mining the structural hole in a co-author network is helpful to improve a team’s innovation ability of the scientific research cooperation. Most of the analysis on the structural hole is the unweighted network. In order to study the closeness on the characteristics of structural hole, it focuses on the weighted co-author networks. By introducing the weight into the constraint function of structural hole, and then improving the existing algorithm HIS for structural hole, it proposes a weighted algorithm for structural hole（W_CIHIS algorithm）. With constraint as measurement for mining structural hole spanners and the public datasets, it compares the W_CIHIS algorithm and the original one. The experimental results show that W_CIHIS algorithm works better.

Key words: structural hole, co-author network, weighted network, constraint

摘要： 科研合作日趋频繁，形成了大量复杂的网络结构，如引文网络、共引网络及合著网络等，合著现象也越来越多。结构洞的发现有助于科研合作中提高团队整体的创新能力。目前对于合著网络的结构洞分析多为无权网络，为了研究合著者的紧密程度在结构洞分析中的作用，以加权的合著网络为研究对象，将权重引入到约束度中，并对现有的结构洞算法——HIS算法进行改进，提出一种基于加权约束度的结构洞挖掘算法（W_CIHIS算法）。用约束度作为衡量指标，将W_CIHIS算法与HIS算法应用到公共数据进行对比，实验结果表明提出的算法有很好的效果。

关键词: 结构洞, 合著网络, 加权网络, 约束度