基于广义后缀树的二分网络社区挖掘算法

发布时间：2018-10-26 11:28

【摘要】：近年来,二分网络的社区挖掘问题得到了极大的关注。提出了一种基于广义后缀树的二分网络社区挖掘算法。首先从二分网络的邻接矩阵中提取网络中每个节点的链接节点序列,然后构建广义后缀树。广义后缀树的每个节点表示二分网络的一个完全二分团,由此获取并调整完全二分团。通过引入二分团的紧密度得到初始的社区划分,最后再对孤立点进行处理以得到最终的社区划分。所提算法不仅能发现重叠社区,而且能得到一对多关系的社区。在人工数据集和真实数据集上的实验表明,所提算法能准确地识别二分网络中的社区个数,获得很好的划分效果。
[Abstract]:In recent years, the community mining problem of bipartite network has been paid great attention to. A community mining algorithm based on generalized suffix tree is proposed in this paper. Firstly, the linked node sequence of each node in the network is extracted from the adjacency matrix of the binary network, and then the generalized suffix tree is constructed. Each node of the generalized suffix tree represents a complete dichotomy of a bipartite network, from which a complete dichotomy is obtained and adjusted. The initial community division is obtained by introducing the compactness of dichotomies, and then the isolated points are treated to obtain the final community division. The proposed algorithm can not only find overlapping communities, but also obtain one-to-many communities. Experiments on artificial data sets and real data sets show that the proposed algorithm can accurately identify the number of communities in the binary network and obtain a good partition effect.
【作者单位】：金陵科技学院信息化建设与管理中心;扬州大学信息工程学院;南京大学计算机软件新技术国家重点实验室;南京航空航天大学计算机科学与技术学院;
【基金】：国家自然科学基金项目(61379066) 江苏省高校自然科学基金项目(15KJD520008) 江苏省现代教育技术研究重点课题(2017-R-54927)资助
【分类号】：O157.5

【相似文献】