Unsupervised land-use change detection using multi-temporal POI embedding

AbstractRapid land-use change detection (LUCD) is pivotal for refined urban planning and management. In this paper, we investigate LUCD through learning embeddings of points of interest (POIs) from multiple temporalities. There are several prominent challenges: (1) the co-occurrence problem of multi-temporal POIs, (2) the heterogeneity of POI categorization, and (3) The lack of human-crafted labels. Therefore, multi-temporal POIs need to be aligned in the embedding space for effective LUCD. This study proposes a multi-temporal POI embedding (MT-POI2Vec) technique for LUCD in a fully unsupervised manner. In MT-POI2Vec, we first utilize random walks in POI networks to capture their single-period co-occurrence patterns; then, we leverage manifold learning to capture (1) single-period categorical semantics of POIs to enforce semantically similar POI embedding to be close and (2) cross-period categorical semantics to align multi-temporal POI embedding in a unified embedding space. We conducted experiments in Shenzhen, China, which demonstrates that the proposed method is effective. Compared with several baseline models, MT-POI2Vec can better align multi-temporal POIs and thus achieve higher performance in LUCD. In addition, our model can effectively identify areas with unchanged land use and land use changes in residential and industrial areas at a fine scale.Keywords: Land-use changeembedding space alignmentpoints of interestPOI embedding AcknowledgementsWe would like to acknowledge the comments and insights from the editors and three anonymous reviewers that helped lift the quality of the article.Disclosure statementNo potential conflict of interest was reported by the author(s).Data and codes availability statementWe share the codes and the sub-sampled data of the study at https://doi.org/10.6084/m9.figshare.24081699.Additional informationFundingThis work was supported by the National Key Research and Development Program of China [2019YFB2102903], the National Natural Science Foundation of China [41801306, 42101421 and 42171466]; the “CUG Scholar” Scientific Research Funds at China University of Geosciences (Wuhan) [2022034], a grant from Alibaba Innovative Research Project [20228670], a Guangdong-Hong Kong-Macau Joint Laboratory Program [2020B1212030009], and a grant from State Key Laboratory of Resources and Environmental Information System. W.H. acknowledges the financial support from the Knut and Alice Wallenberg Foundation.Notes on contributorsYao YaoYao Yao is a professor at China University of Geosciences (Wuhan), a researcher from the Center for Spatial Information Science at the University of Tokyo, and a visiting scholar at Alibaba Group. His research interests are geospatial big data mining, analysis, and computational urban science.Qia ZhuQia Zhu is a graduate student at China University of Geosciences (Wuhan). His research interests are spatial representation learning and urban land use change detection.Zijin GuoZijin Guo is a graduate student at China University of Geosciences (Wuhan). His research interests are trajectory data mining and complex network analysis.Weiming HuangWeiming Huang received his PhD in Geographical Information Science at Lund University, Sweden in 2020. He is a Wallenberg-NTU Postdoctoral Fellow at Nanyang Technological University, Singapore. His research interests mainly include spatial data mining and geospatial knowledge graphs.Yatao ZhangYatao Zhang is a doctoral student at the Mobility Information Engineering lab at ETH Zurich and the Future Resilient Systems at the Singapore-ETH centre. His research interests lie in context-based spatiotemporal analysis, geospatial big data mining, and traffic forecasting.Xiaoqin YanXiaoqin Yan is currently a Ph.D. student in GIScience at the Institute of Remote Sensing and Geographical Information Systems, Peking University, Beijing. His research interests are spatiotemporal big data computing and social perception.Anning DongAnning Dong is a graduate student at China University of Geosciences (Wuhan). His research interests are spatiotemporal big data mining and crime geography.Zhangwei JiangZhangwei Jiang is a staff algorithm engineer at Alibaba Group. His research interests are LBS data mining and research&recommendation algorithm.Hong LiuHong Liu is a senior staff algorithm engineer at Alibaba Group. His research interests are data mining and research&recommendation algorithm.Qingfeng GuanQingfeng Guan is a professor at China University of Geosciences (Wuhan). His research interests are high-performance spatial intelligence computation and urban computing.