多维邻近下浙江城市创新网络演化及其机制研究

Urban Innovation Network of Zhejiang From the Perspective of Multidimensional Proximities

以浙江90个县级城市(城区)为对象,基于大数据专利文本挖掘析取城市间合作申请专利数,构建创新网络,采用空间网络模型和负二项回归模型分析2007—2017年浙江城市创新网络的网络结构、时空演化及创新合作强度的影响机制。研究发现:①浙江城市创新网络规模逐渐扩张,“小世界性”显著,通达性较好,整体呈以杭州湾区为核心的“网络局部化、辐射中心化”特征,等级层次性清晰;②浙江城市创新合作强度与经济规模、教育水平、政策支持、技术势差、城市行政等级、技术邻近性、边界相邻效应显著正相关且受地理距离约束,认知邻近性和制度邻近性与创新合作间分别呈“U”型和倒“U”型曲线关系,网络效应更多在整体网层面上促进了创新合作。在特定省份县级城市层面探讨了如何加强城市创新网络协同效应,以促进地方城市间的创新联系。

Taking ninety county-level cities in Zhejiang Province as geographical units,this article uses spatial analysis and negative binomial regression model to examine the structure,spatio-temporal evolution and impact mechanism of Zhejiang city innovative network from 2007 to 2017,with the data of co-application patents based on big text data mining.The study finds that:Firstly,the scale continues to expand and the structure becomes more and more complex in network evolution.In the study period,the scale of Zhejiang city innovative network has been growing from a sparse network to a dense network,but it is still underdeveloped.The network acts as small-world feature which is superior to random network,and the whole network has good accessibility and communication efficiency.Innovative network shows core-periphery pattern which is centralized,localized,and hierarchical,centers around Hangzhou Great Bay Area,leaves western Zhejiang cities in the edge.Secondly,as for impact mechanism,proximity of different dimension has different impact on the level of city innovative cooperation.In addition,city characteristics and network effects also have significant impacts.The intensity of city innovative cooperation in Zhejiang Province is significantly positively correlated with their economic scale,education level and policy support,and the technological gap and the administrative level also play a positive role.Geographical proximity and border proximity are positive factors affecting innovation cooperation.The relationship between cognitive proximity and innovative cooperation is u-shaped,which indicates that when cognitive proximity is at a low level,it is not conducive to innovation cooperation.However,with the increase of inter-city cognitive proximity,its marginal effect on innovative cooperation becomes more and more strong.On the contrary,relationship of innovative cooperation and institutional proximity shows an inverted‘U-shaped’curve,indicating that institutional proximity has a positive effect on innovative cooperation level at a lower level,but its effect will become weaker with the improvement of city market system.Technology proximity shows a positive effect,indicating that the similarity of technology structure makes the knowledge interaction among the cities more smoothly.Third,network effects play a positive role in promoting intercity innovative cooperation.Compared with individual node network characteristics,network structure characteristics formed by cities surrounding the nodes play a more significant role in promoting cooperation.Innovative network more inclined to show the overall effect,so we should focus on the optimization of network structure,taking the advantage of role of core node.