恐怖网络中基于节点割集分层的组织结构发现方法

“911”等一系列恐怖事件导致越来越多的国家开始重视恐怖组织网络的研究,但社会网络分析中的一些典型度量,如节点的度、介数、接近度等,不能很好地适应具有高度组织性的社会网络.文中针对层次结构性较强的恐怖组织网络,提出了一种新的节点重要性度量,其综合了节点的全局信息与局部信息,利用割集算法对网络节点进行分层获得节点全局信息,局部信息则直接利用了节点的出/入度,并使用BP神经网络方法对综合度量进行参数优化.针对“911”恐怖事件成员网络实际数据的分析结果验证了这种度量的有效性.

Organizational Structure Discovery in Terrorist Networks Based on Cut Set and Hierarchical Method

A growing number of countries begin to pay more attention to the terrorist network after a series of terrorist attacks, such as 9.11 event. But some typical SNA metrics are not well adapted to highly organized network, such as the degree centrality, betweenness centrality, and closeness centrality, etc. This paper proposes a new measure of node importance suitable for in highly organized terrorist network, and it combines the global information with the local information of nodes. A set algorithm is used to obtain the global information with the network node level. Morever, the local information directly is obtained by the node output/input degree. In addition BP neural network method is used to optimize the integrated measurement parameters weights. Experimental results of the 9.11 terrorist network member dataset demonstrate the effectiveness of the new measure.