Effects Of Structural Characteristics Of Complex Knowledge Network On Knowledge Flow

Complex networks cannot only describe the complex self-organized system, but also can be used to model and analyze it. If the"element"and"relationship"are abstracted by the node and edge respectively, we can study text knowledge mining from the point view of complex networks.Complex knowledge networks are typical complex networks. In this paper, we first propose a computational model based on Multi-Agent Based Simulation, to form knowledge networks. Firstly, we investigate the knowledge evolution of different knowledge nodes with different characteristics. The simulation results indicate that agents with high brokerage opportunities or high centrality enhance their knowledge levels more rapidly. Then on the basis of two virtual knowledge networks constructed, we investigate how diverse connecting mechanisms between network A and B differentially influence the co-evolution of the knowledge levels of both networks. The simulation results indicate that knowledge diffusion through different knowledge networks seems to be affected significantly by the characteristics of the agents who build the knowledge connections across these networks. Through purposely building knowledge connections among the agents with high brokerage opportunities or high centrality in network A and network B, the knowledge diffusion through different knowledge networks is the most effective. Finally, we investigate the effects of the density and connecting weights of a less-developed network on knowledge flow between networks. The simulation results indicate that under different connecting mechanisms, the gap between high-developed network and less-developed network is reduced accelerately while network density or connecting weights increase gradually. And the knowledge level of less-developed network could catch up with that of high-developed network under certain network density or connecting weights. Moreover, building connections among the agents with high brokerage opportunities or high centrality in different networks still is the most effective connecting mechanism, despite of different values of network density and connecting weights.The research conclusion of this paper is of great importance to the decision maker, who try to speed up the knowledge level of less-developed knowledge network. Based on the simulation results, we analyzed the Innovation Relay Centre and Silicon Valley-Xin Zhu Knowledge network Alliance. We found that technology agency, government and other organizations play a vital role in the construction, management, maintenance of knowledge network and accelerate the knowledge flow between and in networks.