粤港澳大湾区专利合作申请网络结构及链路预测

目的]基于专利合作申请网络及链路预测,一方面揭示粤港澳大湾区创新网络的结构特征,一方面挖掘科技主体间未来合作的可能性。方法]基于粤港澳大湾区2016—2020年产学研合作申请专利数据构建无向加权网络,分析网络特征;使用AdamicAdar指标对大湾区专利技术合作机会进行链路预测。该方法能够明确现有合作网络情况,提升合作预测准确率和效果,更好地进行关联推荐,进一步提升湾区创新效率。结论]湾区城市间合作网络呈现显著的边缘核心结构;湾区内申请人合作网络稀疏,小世界特征不显著;预测结果发现:湾区内产学研合作机会较多,区域内城市间合作机会较少,主要集中在核电领域。

Structure Analysis and Link Prediction of Patent Technology Cooperation Network in Guangdong-Hong Kong-Macao Greater Bay Area

Purpose]Based on the prediction of patent cooperative application network and link,on the one hand,this paper reveals the structural characteristics of innovation network in Guangdong-Hong Kong-Macao Greater Bay,on the other hand,it explores the possibility of future cooperation between science and technology subjects.Method]Based on the patent application data of industry-universityresearch cooperation in Guangdong-Hong Kong-Macao Greater Bay from 2016 to 2020,the undirected weighted network was constructed,and the network characteristics were analyzed;the adamicadar index was used to predict the patent technology cooperation opportunities in Greater Bay.This method can clarify the existing cooperation network,improve the accuracy and effect of cooperation prediction,make better association recommendation,and further improve the innovation efficiency of the Greater Bay.Conclusion]The inter city cooperation network in the Greater Bay presents a significant marginal core structure;the applicant cooperation network in the Greater Bay is sparse,and the small world characteristics are not significant;the prediction results show that there are more opportunities for industry-university-research cooperation in the Greater Bay,and less opportunities for inter City cooperation in the region.