Structural Analysis And Core-Subnet Extraction Algorithm Of The Complex Trade Network

In the real world, many systems are in the form of complex networks, one of them is the world trade network. World trade network is one of time-varying weighted network, and it is formed by the trade links between different countries. So it is very important that analysis the characteristics of the network topology and sub network and the set of core node, which is of great significance for predicting risks of the world trade. The current existing methods which are used to get the core subnet and core node set of the network have several disadvantages. Firstly, there is not a uniform standard for designing the formula which is used to measure the importance of the node. Secondly, the standard for the size of the core subnet and core node set is not unique. So in this thesis, we use the world trade network as the analysis object, propose a new algorithm to get the core net and core node set of the complex weighted network.Major findings of the current study are recapped as follows:Firstly, this thesis constructs a variety of different types of the world trade network from vast amount of actual international trade data, and introduces the different kinds of network construction in detail, and then analyzes characteristics of several kinds of trade network’s topological structure and community. Moreover, the thesis presents a sorting method which based on the Borda Count algorithm to get multilayer node core net nodes rank.Secondly, this thesis proposes a new algorithm to get weighted network core net nodes. This algorithm uses a sliding W to reach a steady state for the changes of the net, and then get the core net and core node set. The thesis also put forward an approach based on statistical method to get the weighted network core node set. This thesis calculates the frequency of the nodes appearing in the core subnets to identify the importance of the node. At last, this thesis implements a large number of experiments on many kind of trade network, and compares the experiment results with the world trade report given by WTO to validate the validity and correctness of the algorithm in this paper.