21 APR 2026 (TUE) 10:30-12:00

2 hours ago
2 min read

Geography Distinguished Seminars Series

Using NLP to identify urban development typologies in Chinese cities

Date: 21 APR 2026 (Tuesday)

Time: 10:30-12:00 (HKT)

Venue: CLL, Department of Geography, 10/F, The Jockey Club Tower, Centennial Campus, HKU

Mode: Hybrid

Via Zoom: Zoom link will be provided upon successful registration

Registration link: https://hkuems1.hku.hk/hkuems/ec_hdetail.aspx?guest=Y&ueid=106260

Abstract:

This study applies natural language processing (NLP) to analyze the 12th Five-Year Plans (2011–2015) of 286 Chinese prefectural-level and above cities, with two objectives: (1) to understand how Chinese cities differ in their development strategies and socioeconomic profiles; and (2) to assess the potential of computational text-analysis methods for large-scale, systematic analysis of formal planning documents. Using embedding-based NLP model (Doc2Vec) and k-means clustering, we identify six clusters of cities with distinct socio-economic profiles and stated development priorities. We then benchmark these NLP-derived clusters against the human-coded typology reported by Xu and Heikkila (2020). A chi-square test shows a statistically significant association between the two classifications, and correlations between cluster-level socio-economic profiles indicate that the distinction between the NLP-based clusters are at least as evident—and in some respects sharper—than the human-derived ones. Topic modelling with BERTopic further clarifies the main development themes in each cluster, culminating in six development strategies that structured China's urbanization during the 12th Five-Year Plan period. The findings contribute to the literature on urbanization and governance, planning documents as data, and the use of large-scale text-analysis methods for constructing indicators and typologies of urban development.

Professor Eric Heikkila

Professor Emeritus of Public Policy, University of Southern California

Eric Heikkila is Professor Emeritus of Public Policy at the University of Southern California. His primary field of research is urban and regional development in a global context. As a scholar, Professor Heikkila's approach is comprehensively interdisciplinary, combining a range of social science and cultural perspectives, including spatial analytics, economics, cultural history and public policy. As Director of Global Engagement for the USC Price School of Public Policy for fifteen years, Dr. Heikkila fostered collaboration with institutions in Europe, Latin America, and especially East Asia. These ties were further enriched through many visiting appointments in the Asia Pacific region, including a one-year appointment as Vice-Chancellor's Distinguished Visiting Professor at the Chinese University of Hong Kong, 2023-2024. He speaks French and Mandarin to some degree, and has passing familiarity with several other languages.