人工智能驱动的Logistic–Probit建模：粤港澳大湾区制度合作与创新协作识别方法研究

吕 浩; 郭婧颖; 钟雪芬; 周思绮; 刘晓颖

doi:10.63944/essf.AIA

AI-Driven Logistic–Probit Modeling: A Study on Identifying Institutional Cooperation and Innovation Collaboration in the Guangdong-Hong Kong-Macao Greater Bay Area

Hao Lv (Author)

Guangdong University of Education, Guangzhou 510800, Guangdong province, China

Jingying Guo (Author)

Guangdong University of Education, Guangzhou 510800, Guangdong province, China

Xuefen Zhong (Author)

Guangdong University of Education, Guangzhou 510800, Guangdong province, China

Siqi Zhou (Author)

Guangdong University of Education, Guangzhou 510800, Guangdong province, China

Xiaoying Liu (Author)

Guangdong University of Education, Guangzhou 510800, Guangdong province, China

DOI:https://doi.org/10.63944/essf.AIA

Keywords:

人工智能, Logistic 回归, Probit 模型, 制度性合作, 协同创新

Published

31-12-2025

Abstract: This study integrates artificial intelligence and econometric modeling to develop an intelligent
framework for identifying intercity collaborative innovation in the Guangdong–Hong Kong–Macao Greater Bay Area (GBA). A key contribution is the construction of an Institutional Cooperation Index (ICI) that quantifies institutional collaboration between cities through a hybrid expert-scoring and machine learning approach, normalized to [0,1] with an optimized threshold of 0.4 for effective cooperation classification. To reduce subjectivity, AI-based normalization and random perturbation simulation are applied, and validation shows strong correlations between the ICI, R&D intensity, and industrial upgrading. Logistic and Probit regressions are employed as AI classifiers to predict collaboration probabilities using features such as economic linkage, institutional synergy, spatial proximity, and industrial complementarity. NLP techniques extract institutional cooperation features from policy documents, and rare-event correction with network-robust validation ensures prediction stability. Results reveal that institutional cooperation significantly enhances intercity innovation, exhibiting threshold and nonlinear effects. Overall, the AI-driven modeling framework bridges institutional design, network interaction, and innovation outcomes, offering a computationally interpretable foundation for optimizing regional cooperation and intelligent policy decisions in the GBA.

References

[1] Yu J, de Jong M, Li P. Urban network and knowledge flow in the Greater Bay Area: evidence from co-patenting[J]. Cities, 2024, 146: 105051.
[2] Zhang Y, Wang S, Li J. Mining policy texts with NLP for innovation collaboration analysis[J]. Information Processing & Management, 2022, 59(5): 102970.
[3] McFadden D. Conditional logit analysis of qualitative choice behavior[C]// Zarembka P. Frontiers in Econometrics. New York: Academic Press, 1974: 105–142.
[4] Hosmer D W, Lemeshow S, Sturdivant R X. Applied Logistic Regression (3rd ed.)[M]. New York: Wiley, 2013.
[5] King G, Zeng L. Logistic regression in rare events data[J]. Political Analysis, 2001, 9(2): 137–163.
[6] Asheim B, Gertler M. The geography of innovation: regional innovation systems[M]// Fagerberg J, Mowery D, Nelson R. The Oxford Handbook of Innovation. Oxford: Oxford University Press, 2005: 291–317.
[7] Wooldridge J M. Econometric Analysis of Cross Section and Panel Data (2nd ed.)[M]. Cambridge, MA: MIT Press, 2010.
[8] 国家发展和改革委员会. 粤港澳大湾区发展规划纲要[S]. 北京: 人民出版社, 2019.
[9] Chen K, Guan J. Measuring the efficiency of China’s regional innovation systems: application of network data envelopment analysis[J]. Regional Studies, 2012, 46(3): 355–377.
[10] 胡志坚, 王洪海. 区域创新能力测度与协同创新模式研究[M]. 北京: 科学出版社, 2015.
[11] Wang C, Lin G C S. Dynamics of innovation in the Pearl River Delta, China: technological catch-up, relational assets and global linkages[J]. Environment and Planning A, 2013, 45(4): 936–953.
[12] 陈广汉, 刘敏. 粤港澳大湾区建设中的制度协同与协同治理[J]. 华南师范大学学报(社会科学版), 2019(6): 5–15.

AI-Driven Logistic–Probit Modeling: A Study on Identifying Institutional Cooperation and Innovation Collaboration in the Guangdong-Hong Kong-Macao Greater Bay Area

How to Cite