Evolution and quantitative evaluation of China’s green port policies: evidence from text mining and text analysis

This paper analyzes 441 of China's green port policies using text mining and quantitative evaluation methods to promote the sustainable development of China’s ports and accelerate the achievement of the “dual carbon” goal. The study first reviews the evolution of these policies, categorizing them into three stages: the embryonic stage, the startup stage, and the development stage. High-frequency words were extracted for each stage, followed by dimensionality reduction using Principal Component Analysis (PCA) and clustering analysis to classify the policies. The Latent Dirichlet Allocation (LDA) topic model was then applied to identify the main policy themes, and a Policy Modeling Consistency (PMC) index model was developed. The study evaluated 19 representative policies using cosine similarity and correlation coefficients. The analysis revealed several key issues: (1) the entity responsible for policy issuance is singular, with insufficient interdepartmental collaboration; (2) short-term policies are lacking, with weak responsiveness; (3) policies targeting shipping companies are underdeveloped; and (4) there is an imbalance in the use of policy instruments, with a focus on mandatory and hybrid tools. Based on these findings, the paper proposes several suggestions for improving green port development policies.