加权社会网络中重要节点发现算法

发布时间：2018-06-24 15:05

本文选题：社会网络 + 重要节点　；参考：《计算机应用》2013年06期

【摘要】：从社会网络中发现重要节点是一个很有意义的研究问题,目前多数重要节点发现方法是基于不加权网络。由于在社会网络中,节点之间的关系具有强弱差异,社会网络本质上是一个加权网络。对于加权社会网络中的重要节点发现较少有研究。利用节点交互,提出了节点间关系强度的一种度量方法,该方法考虑了节点局部有向交互特征与全局交互特征。利用节点的行为特征定义了节点活跃度。采用关系强度作为边的权重,活跃度作为节点权重形成了加权社会网络。基于PageRank算法的思想,提出了两个改进算法,算法采用节点权值作为阻尼系数,在迭代式过程用边的权重代替了PageRank算法中的入边和。分别选择国内外具有代表性的2个社交网络上的数据集进行大量实验,并分别选择了不同的方法作为比较,实验结果表明改进算法能较好地发现加权社会网络中的重要节点。
[Abstract]:Finding important nodes from social networks is a significant research problem. At present, most important node discovery methods are based on unweighted networks. Social network is essentially a weighted network because of the difference of the relationship between nodes in social network. There is little research on the discovery of important nodes in weighted social networks. A method to measure the strength of the relationship between nodes is proposed by using node interaction. This method takes into account the local and global interaction characteristics of nodes. The node activity is defined by the behavior characteristics of nodes. The relation strength is used as the weight of the edge and the activity degree as the weight of the node to form the weighted social network. Based on the idea of PageRank algorithm, two improved algorithms are proposed. The weight of nodes is used as damping coefficient. In the iterative process, the weight of edges is used to replace the sum of edges in the PageRank algorithm. A large number of experiments were carried out on the data sets of two representative social networks at home and abroad, and different methods were selected for comparison. The experimental results show that the improved algorithm can better find important nodes in weighted social networks.
【作者单位】：北京工商大学计算机与信息工程学院;
【基金】：国家自然科学基金资助项目(61170112) 北京市属高等学校科学技术与研究生教育创新工程建设项目(PXM2012_014213_000037)
【分类号】：TP393.0

【相似文献】