Decoding China's new-type industrialization: Insights from an XGBoost-SHAP analysis

Abstract

Integrating machine learning and explainable artificial intelligence offers superior interpretability for indicator selection compared with traditional evaluation models. This study utilizes the XGBoost-SHAP model to assess the influencing factors, development trends, and spatiotemporal evolution of China's new-type industrialization. The results demonstrate that digitalization emerges as the most critical factor, contributing 41.23% to the weight of indicators, followed by innovation drive at 34.31%, greenization at 16.38%, and aspects of servitization and fusion. China's new industrialization level shows a positive trend, with an accelerated growth rate post-2020, despite significant regional disparities in industrial indices. A spatiotemporal evolution identified a three-tier regional distribution: the eastern coastal, central plain, and western plateau regions. Notably, the improvement in new-type industrialization has expanded from coastal to inland regions, with significant progress observed in the southwestern region, represented by Sichuan Province. Additionally, the analysis revealed that China's level of new industrialization is converging toward the Hu Huanyong Line, suggesting that it may be influenced by population distribution density and natural environmental factors. The research results provide a scientific basis and decision-making guidance for evaluating the dynamics of China's new industrialization.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 39 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.