Research On The Relationship Between Structure And Function Of Complex Technology Trading Network

Technology trading refers to the process of spreading science and technology through human activities.As it tends to be large-scale,it forms an intertwined technology trading network.However,most of the researches from the perspective of complex network only rest on two relatively independent aspects of structure and function,that is,the structural characteristics of network and the application of network functions.There is a lack of in-depth research on the relationship between the two.This paper is based on the transaction data of 2003?2013 technology as the research object,the time series mapping technology trading to complex network,combined with sequential SARIMA analysis method and technology to predict the trading volume of the stepwise regression model,provide guidance for technology market activities,is conducive to the technology market management and macro-control.The research mainly includes two innovations:a new time series forecasting method based on complex network and a research on the relationship between structure and function of technology trading network.Specific work as follows:(1)Study the theory of the framework.Study the basic theory of complex network structure and function;model determination and prediction methods of time series analysis and regression methods.(2)Build a complex technology trading network.Combining graph theory with network knowledge,using the method of constructing network in graph theory,the subject of transaction in technology transaction is mapped to node of network,the transaction relationship of transaction record is mapped to the edge of network,and technology trading network model and data visual display,to explore the trend of technology transactions in different periods of time characteristics.(3)Build the network topology feature time series model and predict.Combining the time series,the SARIMA model of each network topology structure index is fitted separately,and the modeling steps and the characteristics used in the prediction analysis are elaborated,and the prediction of the technical transaction network structure is realized.(4)Build the regression model of the technology trade and predict.Using stepwise regression analysis method,through multi-variable transformation of variables and interaction item analysis,a regression model of trading volume and topological structure index was constructed,in which the point inference,edge inversion and node number had a significant impact on the trading volume.Finally,compared with the time series forecasting model of simple technology transaction,the results show that the combination of complex network forecasting method has higher prediction accuracy.Based on the empirical analysis,this paper presents a time series regression prediction model combined with complex networks and confirms the close correlation between the structure and function of technology trading network.