Key Node Detection in Financial Complex Network

With the development of the financial sector, the growing complexity of financial transaction network, effectively identify a trading network of key trader has a important significance. Trading network abstraction for complex networks, traders abstraction for nodes, trandings between traders abstraction for the edges. The method of degree centrality, clustering coefficient, betweenness centrality, closeness centrality and the like is not sufficient to evaluate the importance of the node. Therefore, we propose a novel algorithm to evaluate the importance of nodes in undirected and unweighted network. We take the degree centrality and clustering coefficient of the nodes as the evaluation indicators, and combine the importance contribution of the nearest and the next nearest nodes. Through normalization and averaging, the benchmark ranking of node importance is obtained, which comprehensively considers the global and local features of the nodes. We used the real trading network data from Zhengzhou Commodity Exchange (ZCE) to conduct three comparative experiments and analyses. The experiment results show that our method has achieved better results, and can effectively identify key trading traders in ZCE.