Identifying urban functional zones using public bicycle rental records and point-of-interest data

Xiaoyi Zhang, WenWen Li, Feng Zhang, Renyi Liu, Zhenhong Du

Research output: Contribution to journalArticlepeer-review

48 Scopus citations


Human mobility data have become an essential means to study travel behavior and trip purpose to identify urban functional zones, which portray land use at a finer granularity and offer insights for problems such as business site selection, urban design, and planning. However, very few works have leveraged public bicycle-sharing data, which provides a useful feature in depicting people’s short-trip transportation within a city, in the studies of urban functions and structure. Because of its convenience, bicycle usage tends to be close to point-of-interest (POI) features, the combination of which will no doubt enhance the understanding of the trip purpose for characterizing different functional zones. In our study, we propose a data-driven approach that uses station-based public bicycle rental records together with POI data in Hangzhou, China to identify urban functional zones. Topic modelling, unsupervised clustering, and visual analytics are employed to delineate the function matrix, aggregate functional zones, and present mixed land uses. Our result shows that business areas, industrial areas, and residential areas can be well detected, which validates the effectiveness of data generated from this new transportation mode. The word cloud of function labels reveals the mixed land use of different types of urban functions and improves the understanding of city structures.

Original languageEnglish (US)
Article number459
JournalISPRS International Journal of Geo-Information
Issue number12
StatePublished - Nov 27 2018


  • Human mobility
  • K-means
  • Land use
  • Topic modelling
  • Traffic analysis zones

ASJC Scopus subject areas

  • Geography, Planning and Development
  • Computers in Earth Sciences
  • Earth and Planetary Sciences (miscellaneous)


Dive into the research topics of 'Identifying urban functional zones using public bicycle rental records and point-of-interest data'. Together they form a unique fingerprint.

Cite this