Research And Realization On The Multi-dimensional Dynamic Social Network Analysis Of The Weapon Trade Relations

Weapons trade has a complex system characteristics related to the characteristicsof the weapon type, manufacturer, exporter, and volume of transactions, and multirelationships between the elements. The traditional static analysis methods can onlyfocus on one or two elements relationship, unable to meet the network analysis need ofweapons trade.In this paper, a complex meta-network model of the weapons trade is provided as anew reference method, we also describe the modeling process and analysis method. Onthe basis of the weapons trade meta-network, we use Taiwan s import arms annual datafrom1992to2008as the sample to establish a complex dynamic network which iscomposed by14sub meta-networks. We analysis the relevant characteristics of theweapons trade network through the timing dynamic, comprehensively evaluate theworld trade annual features of major arms exporting and importing countries using thesocial network analysis indicators such as network density, centrality, cohesion. Finally,we design the weapons trade network analysis system which is composed by weaponsdata online access modules, the weapons trade network analysis module, the deepapplication modules as well as the weapons trade characteristics visualization module.The analysis system provides arms trade data acquisition, analysis, monitoring and otherfunctions.This paper studies deal with over simplified limitations about the traditionalanalysis, expand modeling method for multi relationships between a variety elements,and add dynamic considerations. We use the network indicators to measure the annualcharacteristics, arms trade zoning and so on. The weapons trade network isdeconstructed through multi-dimensional, multi-level, and multi-modal. The abundanceinformation such as the network structure s evolution and interactive characteristicchanges is apparent. We can also identify the associated structural features whichimplicit in the relational data, mine the characteristics of the various nodes behaviorpatterns, find the frequently interact core subnet trade.