The Research On Methods For Detecting Dynamic Change Of Terrorism Network Based On SPC Control Chart

Detecting dynamic change of terrorism network is an important direction in the research of counter-terrorism. It is an important link in the early warning of the terrorism attack by rapidly and accurately recognizing the changes of the network by using SPC control chart to track the changes of the network's measures. This thesis studies the method to transform the terrorism organization into multi-meta network, and transform the huge amount of measures into a global parameter which is a reasonable representative of the network. Then we use dynamic network analysis to get the time series of the global parameters and SPC control chart to monitor their abnormal changes. Finally we compare the results of different control charts to get the reasonable conclusion. The main contents of the thesis include:1. Establish the multi-meta network of the terrorism organization and analyze its longitudinal feature. We use multi-meta network instead of traditional social network to model the terrorism organization. Meanwhile, we build several networks over time and analyze their longitudinal feature to have a better understanding of terrorism organization. This content builds the foundation for the theory and application of change detection in terrorism network.2. Develop the technique for change detection in terrorism network. On the basis of understanding the characteristics and use environment of the traditional SPC, CUSUM and EWMA control chart, we get their feasibility in change detection of the terrorism networks. By extracting the feature of multi-meta network of the terrorism organization, we use Principal Component Analysis(PCA) to establish a global parameter, which can both represent the network as a whole and can be used by SPC control charts to detect changes. Then the results of different control charts are analyzed to determine the network's most likely change time.3. Use Al-Qaeda network for example to empirical analysis. After getting the intelligence of Al-Qaeda organization, we construct its time series networks and use the technique developed in this thesis step by step to detect the changes. We have the Al-Qaeda data set analyzed here to verify the feasibility of the change detection methods proposed in this thesis.4. Analyze the result and make the policy recommendations. We analyze the result of change detection of Al-Qaeda with the actual events occurring during the time. Then we combine the result with China's actual conditions to make feasible policy recommendations for monitoring and early warning of terrorism organization.