A GloVe-Based POI Type Embedding Model for Extracting and Identifying Urban Functional Regions

Points-of-interest (POIs) are an important carriers of location text information in smart cities and have been widely used to extract and identify urban functional regions. However, it is difficult to model the relationship between POIs and urban functional types using existing methods due to insufficient POIs information mining. In this study, we propose a Global Vectors (GloVe)-based, POI type embedding model (GPTEM) to extract and identify urban functional regions at the scale of traffic analysis zones (TAZs) by integrating the co-occurrence information and spatial context of POIs. This method has three main steps. First, we utilize buffer zones centered on each POI to construct the urban functional corpus. Second, we use the constructed corpus and GPTEM to train POI type vectors. Third, we cluster the TAZs and annotate the urban functional types in clustered regions by calculating enrichment factors. The results are evaluated by comparing them against manual annotations and food takeout delivery data, showing that the overall identification accuracy of the proposed method (78.44%) is significantly higher than that of a baseline method based on word2vec. Our work can assist urban planners to efficiently evaluate the development of and changes in the functions of various urban regions.

[1]  Sergei Vassilvitskii,et al.  k-means++: the advantages of careful seeding , 2007, SODA '07.

[2]  Francisco C. Pereira,et al.  Mining point-of-interest data from social networks for urban land use classification and disaggregation , 2015, Comput. Environ. Urban Syst..

[3]  Thomas Blaschke,et al.  Geographic Object-Based Image Analysis – Towards a new paradigm , 2014, ISPRS journal of photogrammetry and remote sensing : official publication of the International Society for Photogrammetry and Remote Sensing.

[4]  C. Karlsson.  Clusters, Functional Regions and Cluster Policies , 2007 .

[5]  Jeffrey Pennington,et al.  GloVe: Global Vectors for Word Representation , 2014, EMNLP.

[6]  Xiaoping Liu,et al.  Sensing spatial distribution of urban land use by integrating points-of-interest and Google Word2Vec model , 2017, Int. J. Geogr. Inf. Sci..

[7]  Christopher Potts,et al.  Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank , 2013, EMNLP.

[8]  Benjamin Adams,et al.  A data-driven approach to exploring similarities of tourist attractions through online reviews , 2018, J. Locat. Based Serv..

[9]  Carme Miralles-Guasch,et al.  Green exposure of walking routes and residential areas using smartphone tracking data and GIS in a Mediterranean city , 2019, Urban Forestry & Urban Greening.

[10]  Yu Shi,et al.  Beyond Word2vec: An approach for urban functional region extraction and identification by combining Place2vec and POIs , 2019, Comput. Environ. Urban Syst..

[11]  Geoffrey Zweig,et al.  Linguistic Regularities in Continuous Space Word Representations , 2013, NAACL.

[12]  Krzysztof Janowicz,et al.  Extracting urban functional regions from points of interest and human activities on location‐based social networks , 2017, Trans. GIS.

[13]  Liangpei Zhang,et al.  Scene Classification Based on the Multifeature Fusion Probabilistic Topic Model for High Spatial Resolution Remote Sensing Imagery , 2015, IEEE Transactions on Geoscience and Remote Sensing.

[14]  Ellen Banzhaf,et al.  Monitoring Urban Land use Changes with Remote Sensing Techniques , 2011 .

[15]  Shougeng Hu,et al.  Automated urban land-use classification with remote sensing , 2013 .

[16]  Felix Klanner,et al.  Modeling hierarchical category transition for next POI recommendation with uncertain check-ins , 2020, Inf. Sci..

[17]  Xia Li,et al.  The Traj2Vec model to quantify residents’ spatial trajectories and estimate the proportions of urban land-use types , 2020, Int. J. Geogr. Inf. Sci..

[18]  Chaogui Kang,et al.  Social Sensing: A New Approach to Understanding Our Socioeconomic Environments , 2015 .

[19]  Gang Xu,et al.  Street Centralities and Land Use Intensities Based on Points of Interest (POI) in Shenzhen, China , 2018, ISPRS Int. J. Geo Inf..

[20]  Sara Irina Fabrikant,et al.  Spatialization Methods: A Cartographic Research Agenda for Non-geographic Information Visualization , 2003 .

[21]  Naixia Mou,et al.  Visualizing and exploring POI configurations of urban regions on POI-type semantic space , 2020 .

[22]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[23]  Xingjian Liu,et al.  Automated identification and characterization of parcels (AICP) with OpenStreetMap and Points of Interest , 2013, ArXiv.

[24]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[25]  Ghim Ping Ong,et al.  Spatial-temporal inference of urban traffic emissions based on taxi trajectories and multi-source urban data , 2018, Transportation Research Part C: Emerging Technologies.

[26]  Martin J. Dürst,et al.  Location2Vec: Generating Distributed Representation of Location by Using Geo-tagged Microblog Posts , 2018, SocInfo.

[27]  Hwee Tou Ng,et al.  Corpus-Based Approaches to Semantic Interpretation in NLP , 1997, AI Mag..

[28]  Yoshua Bengio,et al.  A Neural Probabilistic Language Model , 2003, J. Mach. Learn. Res..

[29]  Zili Zhu,et al.  Real options analysis for land use management: Methods, application, and implications for policy. , 2015, Journal of environmental management.

[30]  Sohrab Rahimi,et al.  Place niche and its regional variability: Measuring spatial context patterns for points of interest with representation learning , 2019, Comput. Environ. Urban Syst..

[31]  Steven Schockaert,et al.  Embedding Geographic Locations for Modelling the Natural Environment using Flickr Tags and Structured Data , 2019, ECIR.

[32]  Xing Xie,et al.  Discovering regions of different functions in a city using human mobility and POIs , 2012, KDD.

[33]  Meng Joo Er,et al.  PCA and LDA in DCT domain , 2005, Pattern Recognit. Lett..

[34]  Hechan Tian,et al.  Heterogeneous graph-based joint representation learning for users and POIs in location-based social network , 2020, Inf. Process. Manag..

[35]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[36]  Zhenhong Du,et al.  GSAM: A deep neural network model for extracting computational representations of Chinese addresses fused with geospatial feature , 2020, Comput. Environ. Urban Syst..

[37]  Jinchao Song,et al.  Monitoring finer-scale population density in urban functional zones: A remote sensing data fusion approach , 2019, Landscape and Urban Planning.

[38]  Licia Capra,et al.  Urban Computing: Concepts, Methodologies, and Applications , 2014, TIST.

[39]  Jung-Hsien Chiang,et al.  Real-time event embedding for POI recommendation , 2019, Neurocomputing.

[40]  An Luo,et al.  Detecting Urban Polycentric Structure from POI Data , 2019, ISPRS Int. J. Geo Inf..

[41]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[42]  Shihong Du,et al.  Semantic Classification of Heterogeneous Urban Scenes Using Intrascene Feature Similarity and Interscene Semantic Dependency , 2015, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[43]  Krzysztof Janowicz,et al.  From ITDL to Place2Vec: Reasoning About Place Type Similarity and Relatedness by Learning Embeddings From Augmented Spatial Contexts , 2017, SIGSPATIAL/GIS.