Cluster analysis and comparison of development modes of Global Innovation Hubs

Abstract

Abstract: Global Innovation Hubs are centers that emerge as a result of the extensive development and geographical diffusion of science and technology innovation activities, serving as crucial nodes within the global innovation network. The development modes of these hubs represent abstract summaries of the historical evolution and practical implementation of regional science and technology innovation systems. Identifying and comparing these modes is an essential starting point for uncovering regional innovation patterns and formulating targeted development strategies. Previous studies have primarily examined Global Innovation Hubs from the perspective of their temporal dynamics in historical evolution and spatial positioning within the global innovation network. This study aims to further enhance our understanding by exploring the diversity of regional innovation modes, thereby unraveling various developmental paths embedded within different points across the global innovation network. Through more than 10 rounds of expert deliberation spanning three years, this paper selects 31 characteristic indicators to assess regional scientific research capability, innovative economy, and innovative ecology. Using the two-stage K-means clustering method, this study identifies and systematically compares the development patterns of 100 global cities (metropolitan areas). The findings reveal that Global Innovation Hubs exhibit ten distinct types of differentiated development modes: all-around champion hub, international financial hub, ecological hub, knowledge hub, data-driven hub, innovation highland hub, developing hub, international communication hub, growing hub, and quality hub. Notably among them are Chinese cities which predominantly fall into three modes: Beijing and the Guangdong-Hong Kong-Macao Greater Bay Area are the all-around champion hubs; Shanghai is the innovation highland hub; while 15 other cities (metropolitan areas) including Hangzhou, Nanjing and Wuhan are the growing hubs. This study primarily focuses on analyzing these three types of Global Innovation Hubs including China cities. It is observed that they possess distinctive characteristics as follows: Firstly, all-around champion hubs attract top scientific talents worldwide along with leading institutions and facilities. They dominate global frontier technological innovations as well as cooperation networks while connecting with a global technology ecosystem subsystem. Moreover they gather high-tech enterprises globally with outstanding entrepreneurial ecological advantages. Secondly, innovation highland hubs emerge as prominent hubs for emerging technologies and industries at a global level, by fostering strong industry-university-research alliances alongside industrial parks formation where government plays an instrumental role in shaping regional industrial clusters. Thirdly, growing hubs are regional economic growth poles, driving and radiating science and technology innovation activities in the region and its surrounding small satellite cities. However, due to limited scientific knowledge accumulation and technological history, these hubs still function as secondary nodes within the global innovation network. The research findings contribute a novel empirical approach and data foundation for the systematic identification and comparison of Global Innovation Hubs. Moreover, the inclusion of emerging cities (metropolitan areas) such as Beijing and the Guangdong-Hong Kong-Macao Greater Bay Area provides compelling evidence to elucidate the evolution of global innovation geography, particularly highlighting China's ascent as an emerging scientific and technological powerhouse. Furthermore, this study enhances the understanding of Global Innovation Hubs by exploring diverse regional innovation modes. The conclusion reveals ten distinct development models that further substantiate how regions can formulate their own "regional formula" for science, technology, innovation, and industrial advancement based on their unique knowledge base and innovation resources. Identifying development modes can assist China's Global Innovation Hubs in accurately positioning their niche and benchmark city within the global innovation network, providing empirical support and decision-making references for establishing Global Innovation Hubs in China.

摘要： 国际科技创新中心发展模式是对区域科技创新制度历史与实践活动的抽象总结。识别与比较国际科技创新中心的发展模式是挖掘区域创新规律、定位区域特色发展战略的必要抓手。本文基于历时三年、多达十余轮的专家论证，遴选了31项衡量区域科技创新能力的特征指标，采用聚类分析，对全球100个城市（都市圈）的发展模式进行识别与系统比对。研究发现，国际科技创新中心具有10类差异化的发展模式，其中我国城市集中分布于3类科技创新中心模式，分别为：全能冠军型、创新高地型和成长型。发展模式的识别帮助我国科技创新中心准确定位在全球创新网络中的生态位与对标城市，为我国建成具有全球影响力的科技创新中心提供实证支撑与决策参考。

陈玲孙君孔文豪汪佳慧. 国际科技创新中心发展模式的聚类分析与比较[J]. 科学学研究, 2024, 42(9): 1967-1978.

References

[1]陈搏.全球科技创新中心评价指标体系初探[J].科研管理,2016,37(S1):289-295. Chen B. A preliminary research on the global science － technology innovation [2]center evaluation indicator system[J]. Science Research Management,2016,37(S1):289-295. [3] 熊鸿儒.全球科技创新中心的形成与发展.[J].学习与探索,2015,242(9):112-116.Xiong H R. The formation and development of Global Science and Technology Innovation Centers[J]. Study & Exploration, 2015, 242(9): 112-116. [4] Yassa M. Center of scientific activity: its shift from the 16th to the 20th century[J]. Japanese Studies in the History of Science,1962,1(1):57-75. [5] 杨从平,郑世珏. 科学中心的转移与我国机遇分析[J]. 学术论坛, 2015, 38(1):148-151.Yang C P, Zheng S J. 2015. Transfer of science centers and analysis of opportunities in China [J] Academic Forum, 2015, 38(1): 148-151. [6] 陈劲,阳银娟.协同创新的理论基础与内涵[J].科学学研究,2012,30(02):161-164. Chen J, Duan Y J. Theoretical basis and content for collaborative innovation[J]. Studies in Science of Science, 2015, 38(1): 148-151. [7] 中央组织部党建研究所课题组.当今世界主要发展模式比较研究(四篇)[J].当代世界与社会主义,2011, 89(1):18-29. Research group of the party building research institute of the central organization department. Comparative study of major development models in the world today (Four Articles) [J]. Contemporary World and Socialism, 2011 ,89(1): 18-29. [8] 陈红花,尹西明,陈劲,等. 基于整合式创新理论的科技创新生态位研究[J]. 科学学与科学技术管理, 2019, 40(5): 3-16. Chen H H, Ying X M, Chen J et al. Research on scientific and technological innovation niche based on holistic innovation[J]. Science of Science and Management of S.& T., 2019, 40(5): 3-16. [9] 杨丹辉. 应加快建设具有全球影响力的科技创新中心[J]. 人民论坛·学术前沿, 2020, 190(6): 23-31. Yang D H. Accelerating the building of technological innovation centers with global influence[J]. Frontiers, 2020, 190(6): 23-31. [10] 陈强,王浩,敦帅. 全球科技创新中心:演化路径、典型模式与经验启示[J]. 经济体制改革, 2020, 222(3): 152-159. Chen Q, Wang H, Dun S. Global science and technology innovation center: evolution path, typical model and experience enlightenment[J]. Reform of economic system, 2020, 222(3): 152-159. [11] Ernst D. A new geography of knowledge in the electronics industry? Asia's role in global innovation networks[R]. Singapore: East-West Center Policy Studies series #54, 2009. 32-37. [12] 杜德斌,段德忠. 全球科技创新中心的空间分布、发展类型及演化趋势[J]. 上海城市规划, 2015, 120(1): 76-81. Du D B, Duan D Z. Spatial Distribution, development type and evolution trend of global science and technology innovation center[J]. Shanghai Urban Planning Review, 2015, 120(1): 76-81. [13] Clark, J., Harrison, J., & Miguelez, E. (2018). Connecting cities, revitalizing regions: the centrality of cities to regional development. Regional Studies, 52(8), 1025–1028. [14] Derudder B , Taylor P J . Central flow theory: comparative connectivities in the world-city network[J]. Regional Studies, 2018, 52(8): 1029-1040. [15] 徐晓丹,柳卸林.中国区域科技创新与发展40年[J].科学学研究,2018,36(12):2136-2140. Xu X D, Liu X L. 40 years of co － evolution of China’s S＆T policy and industries[J]. Studies in Science of Science, 2018,36(12):2136-2140. [16] Sassen S , Princeton University Press. The Global City: New York, London, Tokyo[M]. Princeton University Press, 2013. [17] Csomos G , Toth G . Exploring the position of cities in global corporate research and development: A bibliometric analysis by two different geographical approaches[J]. Journal of Informetrics, 2016, 10(2):516-532. [18] 曹湛,彭震伟. 崛起的全球创新中心:中国城市在全球城市科研合作网络中的演化特征[J]. 城市规划学刊, 2021, 265(5): 23-31. Cao Z, Peng Z W. Rising global innovative centers: The evolution of Chinese cities in global inter-urban scientific collaboration networks[J]. Urban Planning Forum, 2021, 265(5): 23-31. [19]清华大学产业发展与环境治理研究中心和Nature Research. 全球科技创新中心指数2020. 北京：中关村论坛: 2020: 9. [20]清华大学产业发展与环境治理研究中心和Nature Research. 全球科技创新中心指数2020. 北京：中关村论坛: 2021: 2. [21]清华大学产业发展与环境治理研究中心和Nature Research. 国际科技创新中心指数2022[R/OL].[2023-04-20], http://www.gov.cn/gzdt/2006-02/28/content_213129.htm, [2023-04-20]. [22] United Nations. Population Division, The World’s cities in 2016: data booklet[R]. New York: UN，2016, 1-26. [23] 中国社会科学院财经院和与联合国人居署.全球城市竞争力报告2019-2020：跨入城市的世界300年变局[R].北京：中国社会科学院财经战略研究院、联合国人居署，2021. 2-7. The Chinese Academy of Social Sciences (CASS) and UN Habitat jointly. Global city competitiveness report 2019-2020: 300 years of change in the world entering cities [R]. Beijing: Chinese Academy of Social Sciences, UN Habitat, 2021: 2-7. [24] Nature. Nature Index 2018 Science Cities[R/OL]. [2023-04-20]. https://www.nature.com/collections/ydnhjfxrzv.[2023-04-20]. [25] Grossetti, M., Eckert, D., Gingras, Y., Jégou, L., Larivière, V., & Milard, B. (2014). Cities and the geographical deconcentration of scientific activity: A multilevelanalysis of publications (1987–2007). Urban Studies, 51(10), 2219–2234. [26] United Nations, Department of Economic and Social Affairs, Population Division. World Urbanization Prospects: The 2018 Revision (ST/ESA/SER.A/420)[R]. New York: United Nations，2019, 57. [27] Aldenderfer MS, Blashfield RK. Cluster Analysis[M]. Los Angeles: Sage Publications, 1984: 2-12. [28] 章永来,周耀鉴. 聚类算法综述[J]. 计算机应用, 2019, 39(7): 1869-1882. Zhang Y L, Zhou Y J. Ｒeview of clustering algorithms[J]. Journal of Computer Applications, 2019, 39(7): 1869-1882. [29] 文宏.政治获得感评价指标体系与地区比较实证研究——基于因子分析和聚类分析[J].经济社会体制比较,2020, 209(3):96-106. Wen H. An empirical study of the evaluation index system of the political sense of acquisition, and the comparison between the regional political sense of acquisition: based on a factor analysis and cluster analysis[J]. Comparative Economic & Social Systems,2020, 209(3):96-106.