| The innovation resources and capabilities owned by a single subject often cannot meet its own needs,and multi-dimensional,multi-subject collaborative innovation is the key to the improvement of regional innovation capabilities.Compared with developed economic regions,Jiangxi Province and other underdeveloped innovation regions in the central and western regions need to be improved in terms of innovation development balance and innovation synergy.The changes in innovation cooperation in multiple dimensions need urgent attention from scholars.Dependent multilayer network,as a multilayer network with homogeneous nodes within layers and heterogeneous nodes between layers,is a frontier method for studying structural equilibrium and optimization problems in social and economic systems.Therefore,this paper uses the dependent multiplex networks to explore the current situation and development direction of the regional innovation cooperation system in Jiangxi Province from the three dimensions of technical knowledge,innovation subject and innovation city.The main research contents include:First of all,the relevant literature on innovation cooperation and innovation cooperation multilayer network is systematically sorted out,supported by knowledge flow theory,social network theory and regional innovation system theory.Secondly,the knowledge network,subject network and city network in the process of innovative cooperation in Jiangxi Province are analyzed by using the methods of knowledge flow characteristic analysis,social network analysis,network optimal link mechanism and spatial structure analysis.The network structure characteristics,key nodes and knowledge flow characteristics in the knowledge network are expounded in detail;the network structure characteristics,optimal link mechanism,key nodes and structural holes of the main network;and the network structure characteristics,optimal link mechanism and spatial structure in the urban network feature.Thirdly,it clarifies the construction method of multi-layer network of dependent innovation cooperation,and determines the inter-layer structure measurement indicators of multilayer network: knowledge diversity,knowledge uniqueness,subject space radiation degree and urban subject bearing degree,etc.The inter-layer structure analysis of provincial innovation cooperation multi-layer network was carried out.Finally,it summarizes the shortcomings of innovation cooperation in Jiangxi Province,and puts forward corresponding suggestions.The research shows that: in the single-layer network of innovation cooperation,innovation cooperation in Jiangxi Province has not yet achieved the integration and innovation of cross-field technical knowledge;the exchange density between innovation subjects is increasing,and schoolenterprise cooperation is also relatively active,but large-scale innovation has not yet formed Collaborative clusters hinder the flow of knowledge;the stability of urban cooperation continues to increase,and Nanchang’s core role continues,gradually forming an innovative spatial pattern of "single core-double core-edge".relatively weak.In the multi-layer network of innovation cooperation,a "single-center-multi-center" spatial correlation structure is formed between cities and innovation subjects,and a "multi-center-multi-center" spatial correlation structure is formed between innovation subjects and knowledge;Knowledge diversity and uniqueness are both high,and universities,power companies and research institutes,pharmaceutical and medical equipment companies have a wide range of main space radiation;Nanchang’s knowledge diversity is much higher than other prefecture-level cities,followed by Ganzhou,Yichun and Jiujiang,but the regional imbalance of the city’s collaborative innovation capability is more obvious.Finally,it is suggested to optimize the technical knowledge combination structure of regional innovation cooperation,enhance the participation of regional scientific research institutes in innovation cooperation,improve the benign interaction channels between innovation subjects,and build an innovation ecosystem of "one core and two centers". |