基于优化结构洞的无向加权网络关键节点发现方法

发布时间：2018-10-07 17:46

【摘要】：针对目前大多数关键节点发现算法没有兼顾桥节点与其他类型关键节点,造成评价结果存在片面性的问题,使用加权网络模型结合结构洞理论,提出一种优化结构洞的无向加权网络关键节点发现方法。综合考虑了节点的邻居数量及其与邻居间的拓扑结构,首先通过定义节点的邻接度和二次邻接度来衡量邻居节点对其的重要程度,在此基础上测量网络中的结构洞约束系数并通过排序发现网络中处于重要位置的关键节点。该方法既反映出节点局部连接的特性,又可在全局拓扑未知的情况下发现其中的关键节点,解决了全局方法计算复杂度高的问题。实验结果表明,该方法比基于介数、节点强度、接近度方法更准确、有效地发现无向加权网络中的关键节点。
[Abstract]:In order to solve the problem of one-sidedness of evaluation results caused by the fact that most key node discovery algorithms do not take account of bridge nodes and other key nodes, a weighted network model combined with structure hole theory is used. This paper presents an undirected weighted network key node discovery method for optimizing structural holes. The number of neighbors and the topological structure between them are considered synthetically. Firstly, the importance of neighbor nodes to them is measured by defining the degree of adjacency and the degree of quadratic adjacency. On this basis, the structural hole constraint coefficients in the network are measured and the key nodes in the network are found by sorting. This method not only reflects the characteristic of local connection of nodes, but also finds the key nodes under the condition of unknown global topology, which solves the problem of high computational complexity of the global method. The experimental results show that the proposed method is more accurate than the one based on medium, node strength and proximity, and can effectively find the key nodes in undirected weighted networks.
【作者单位】：桂林电子科技大学信息与通信学院;
【基金】：国家科技支撑计划课题资助项目(2014BAK11B02) 广西科学研究与技术开发计划资助项目(桂科攻14122007-5) 桂林电子科技大学研究生教育创新计划资助项目(YJCXS201521)
【分类号】：O157.5

【相似文献】