Research On Cooperative Behavior And Performance Evaluation Of Industry-University-Research Patent Cooperation Network

The Industry-university-research cooperation,as a kind of innovation engine,can not only discover and create new knowledge and technology,but also be an important way to build and accelerate the development of national innovation.Patent is an important carrier for the transformation of scientific and technological achievements,and joint patent application has become an important way for enterprises,universities and scientific research institutes to cooperate in innovation and development.Nowadays,Industry-university-research cooperation has been transformed from a point-to-point linear mode to a networked mode,and the networked development mode has also brought the patent cooperation research into a new stage,forming a patent cooperation network composed of enterprises,scientific research institutes and universities as the subjects.China’s Industry-university-research cooperation has tried a number of exploration and practice,and has made great development,but there are some urgent problems to be solved.For example,many Industry-university-research cooperations in China have a short-term trend,the form of cooperation is relatively loose,the degree of Industryuniversity-research cooperation patent cooperation is low,the number of participating subjects is limited,and the cooperation relationship is unstable,which leads to the subjects of Industryuniversity-research cooperation can’t improve the independent innovation ability,which is not conducive to the creation of new knowledge,new technology and the sharing of knowledge,technology and other resources.Complex network theory provides an efficient research perspective and method for the logical relationship of the structure,cooperative behaviors and performance evaluation of the Industry-university-research patent cooperation network.Based on complex network theory,link prediction theory and network evolution game theory,this study combines the actual situation of Industry-university-research patent cooperation in China,concentrates on the network perspective and adaptes the classical research line of "structure-behavior-performance",based on a comprehensive analysis of the topology of the Industry-university-research patent cooperation network,identification of cooperative relationships and selection of cooperative strategies,this study evaluates the performance of the Industry-university-research patent cooperation network and proposes countermeasures to improve the performance of the Industry-university-research patent cooperation network.First,the theoretical framework of the Industry-university-research patent cooperation network is determined.Starting from the basic concepts of industry-university-research cooperation,Industry-university-research patent cooperation and Industry-university-research patent cooperation network,the basic models of network are summarized,the characteristics of Industry-university-research patent cooperation network and the concept of cooperation behavior are systematically analyzed,the internal logical relationship among the research contents of each part is sorted out,and the research framework of Industry-university-research patent cooperation network is determined ultimately,which lays the foundation for the following research.Secondly,the structure and community discovery of the Industry-university-research patent cooperation network are deeply analyzed.By collecting relevant patent application data of the State Intellectual Property Office of China from 2007 to 2019,the Industry-universityresearch patent cooperation network is constructed,the topological structure of the Industryuniversity-research patent cooperation network is analyzed,and the Industry-universityresearch patent cooperation network possesses the characteristics of a small-world network.Then,using the attractive force between nodes,the community detection method based on the attractive force between nodes is proposed to divide the nodes in the network into corresponding communities,and the structure of the Industry-university-research patent cooperation network is deeply analyzed,the communities where the important nodes located in the network are analyzed.Thirdly,the cooperative relationship in the Industry-university-research patent cooperation network is identified.Network is composed of nodes and edges between nodes.This chapter considers the edges between nodes in the Industry-university-research patent cooperation network,proposes a link prediction algorithm based on attractive force between nodes and community detection,and detects the missing links and spurious links in the university-industry patent cooperation network by calculating the connection probability and break probability between two nodes,the cooperation relationship in the university-industry patent cooperation network is identified.Fourthly,the patent cooperation strategies selection of the subjects of Industry-universityresearch patent cooperation network are analyzed based on the network evolutionary game.Considering the attributes of the nodes in the Industry-university-research patent cooperation network,this study uses “capability” as a quantitative index to measure the innovation level of the subjects of the Industry-university-research patent cooperation network,this study proposes an Industry-university-research patent cooperation network evolution game model integrating the factor of “mutation”,analyzes the revenue matrix of the subjects of the Industry-universityresearch patent cooperation network,expounds the strategy renewal rules in the condition of“complete rationality” and “mutation”,and then the evolution game model of the Industryuniversity-research patent cooperation network containing two stages is constructed.The proposed model is validated on the Industry-university-research patent cooperation network,including the model parameters,the number of subjects corresponding to different strategies,and the evolutionary equilibrium.Fifthly,the performance evaluation of the Industry-university-research patent cooperation network.Based on the above comprehensive analysis of the network structure,cooperative relationship and cooperative strategies of the Industry-university-research patent cooperation network,and on the basis of the design principles of the performance evaluation indexes of the Industry-university-research patent cooperation network,performance evaluation index system of the Industry-university-research patent cooperation network is constructed,the entropy value method-TOPSIS method is used to analyze the empirical results of nine local networks’ network performance in the Industry-university-research patent cooperation network.This study also proposes countermeasures for improving the performance of the Industry-university-research patent cooperation network.