Evolution of Chinese original-innovation talent policies: a topic modelling approach

Abstract

ABSTRACT Original-innovation talent significantly influences a nation’s competitive edge. Consequently, supportive policies towards original-innovation talents are critical for the enhancement of national innovation governance. This study utilised a Latent Dirichlet Allocation (LDA) model to conduct a comprehensive analysis of the development and evolution of 334 original-innovation talent policies enacted in China from 1978 to 2022, where we scrutinised the features and progression of China’s original-innovation talent policies and recapitulated the course of their maturation. Semantic analysis was employed to pinpoint the keyword characteristics of these policies, while the LDA model was utilised to extract relevant information, study the evolution of topic intensity, and dissect the fluctuation of topic intensity across various periods. The results indicated that China’s original-innovation talent policies have metamorphosed in alignment with the nation's development objectives, which now focus predominantly on the support of breakthroughs in key core technologies and basic research. The paper concludes with suggestions for strengthening the top-level design of policies, optimising incentive policies, and creating a supportive environment for original innovation.