Unsupervised representation learning of spatial data via multimodal embedding

Porter Jenkins, Ahmad Farag, Suhang Wang, Zhenhui Li

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Increasing urbanization across the globe has coincided with greater access to urban data; this enables researchers and city administrators with better tools to understand urban dynamics, such as crime, traffic, and living standards. In this paper, we study the Learning an Embedding Space for Regions (LESR) problem, wherein we aim to produce vector representations of discrete regions. Recent studies have shown that embedding geospatial regions in a latent vector space can be useful in a variety of urban computing tasks. However, previous studies do not consider regions across multiple modalities in an end-to-end framework. We argue that doing so facilitates the learning of greater semantic relationships among regions. We propose a novel method, RegionEncoder, that jointly learns region representations from satellite image, point-of-interest, human mobility, and spatial graph data. We demonstrate that these region embeddings are useful as features in two regression tasks and across two distinct urban environments. Additionally, we perform an ablation study that evaluates each major architectural component. Finally, we qualitatively explore the learned embedding space, and show that semantic relationships are discovered across modalities.

Original languageEnglish (US)
Title of host publicationCIKM 2019 - Proceedings of the 28th ACM International Conference on Information and Knowledge Management
PublisherAssociation for Computing Machinery
Pages1993-2002
Number of pages10
ISBN (Electronic)9781450369763
DOIs
StatePublished - Nov 3 2019
Event28th ACM International Conference on Information and Knowledge Management, CIKM 2019 - Beijing, China
Duration: Nov 3 2019Nov 7 2019

Publication series

NameInternational Conference on Information and Knowledge Management, Proceedings

Conference

Conference28th ACM International Conference on Information and Knowledge Management, CIKM 2019
CountryChina
CityBeijing
Period11/3/1911/7/19

Fingerprint

Urbanization
Crime
Standard of living
Urban environment
Urban dynamics
Graph
Globe

All Science Journal Classification (ASJC) codes

  • Business, Management and Accounting(all)
  • Decision Sciences(all)

Cite this

Jenkins, P., Farag, A., Wang, S., & Li, Z. (2019). Unsupervised representation learning of spatial data via multimodal embedding. In CIKM 2019 - Proceedings of the 28th ACM International Conference on Information and Knowledge Management (pp. 1993-2002). (International Conference on Information and Knowledge Management, Proceedings). Association for Computing Machinery. https://doi.org/10.1145/3357384.3358001
Jenkins, Porter ; Farag, Ahmad ; Wang, Suhang ; Li, Zhenhui. / Unsupervised representation learning of spatial data via multimodal embedding. CIKM 2019 - Proceedings of the 28th ACM International Conference on Information and Knowledge Management. Association for Computing Machinery, 2019. pp. 1993-2002 (International Conference on Information and Knowledge Management, Proceedings).
@inproceedings{ebd016ce619b47a194896ed55c2c4f27,
title = "Unsupervised representation learning of spatial data via multimodal embedding",
abstract = "Increasing urbanization across the globe has coincided with greater access to urban data; this enables researchers and city administrators with better tools to understand urban dynamics, such as crime, traffic, and living standards. In this paper, we study the Learning an Embedding Space for Regions (LESR) problem, wherein we aim to produce vector representations of discrete regions. Recent studies have shown that embedding geospatial regions in a latent vector space can be useful in a variety of urban computing tasks. However, previous studies do not consider regions across multiple modalities in an end-to-end framework. We argue that doing so facilitates the learning of greater semantic relationships among regions. We propose a novel method, RegionEncoder, that jointly learns region representations from satellite image, point-of-interest, human mobility, and spatial graph data. We demonstrate that these region embeddings are useful as features in two regression tasks and across two distinct urban environments. Additionally, we perform an ablation study that evaluates each major architectural component. Finally, we qualitatively explore the learned embedding space, and show that semantic relationships are discovered across modalities.",
author = "Porter Jenkins and Ahmad Farag and Suhang Wang and Zhenhui Li",
year = "2019",
month = "11",
day = "3",
doi = "10.1145/3357384.3358001",
language = "English (US)",
series = "International Conference on Information and Knowledge Management, Proceedings",
publisher = "Association for Computing Machinery",
pages = "1993--2002",
booktitle = "CIKM 2019 - Proceedings of the 28th ACM International Conference on Information and Knowledge Management",

}

Jenkins, P, Farag, A, Wang, S & Li, Z 2019, Unsupervised representation learning of spatial data via multimodal embedding. in CIKM 2019 - Proceedings of the 28th ACM International Conference on Information and Knowledge Management. International Conference on Information and Knowledge Management, Proceedings, Association for Computing Machinery, pp. 1993-2002, 28th ACM International Conference on Information and Knowledge Management, CIKM 2019, Beijing, China, 11/3/19. https://doi.org/10.1145/3357384.3358001

Unsupervised representation learning of spatial data via multimodal embedding. / Jenkins, Porter; Farag, Ahmad; Wang, Suhang; Li, Zhenhui.

CIKM 2019 - Proceedings of the 28th ACM International Conference on Information and Knowledge Management. Association for Computing Machinery, 2019. p. 1993-2002 (International Conference on Information and Knowledge Management, Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Unsupervised representation learning of spatial data via multimodal embedding

AU - Jenkins, Porter

AU - Farag, Ahmad

AU - Wang, Suhang

AU - Li, Zhenhui

PY - 2019/11/3

Y1 - 2019/11/3

N2 - Increasing urbanization across the globe has coincided with greater access to urban data; this enables researchers and city administrators with better tools to understand urban dynamics, such as crime, traffic, and living standards. In this paper, we study the Learning an Embedding Space for Regions (LESR) problem, wherein we aim to produce vector representations of discrete regions. Recent studies have shown that embedding geospatial regions in a latent vector space can be useful in a variety of urban computing tasks. However, previous studies do not consider regions across multiple modalities in an end-to-end framework. We argue that doing so facilitates the learning of greater semantic relationships among regions. We propose a novel method, RegionEncoder, that jointly learns region representations from satellite image, point-of-interest, human mobility, and spatial graph data. We demonstrate that these region embeddings are useful as features in two regression tasks and across two distinct urban environments. Additionally, we perform an ablation study that evaluates each major architectural component. Finally, we qualitatively explore the learned embedding space, and show that semantic relationships are discovered across modalities.

AB - Increasing urbanization across the globe has coincided with greater access to urban data; this enables researchers and city administrators with better tools to understand urban dynamics, such as crime, traffic, and living standards. In this paper, we study the Learning an Embedding Space for Regions (LESR) problem, wherein we aim to produce vector representations of discrete regions. Recent studies have shown that embedding geospatial regions in a latent vector space can be useful in a variety of urban computing tasks. However, previous studies do not consider regions across multiple modalities in an end-to-end framework. We argue that doing so facilitates the learning of greater semantic relationships among regions. We propose a novel method, RegionEncoder, that jointly learns region representations from satellite image, point-of-interest, human mobility, and spatial graph data. We demonstrate that these region embeddings are useful as features in two regression tasks and across two distinct urban environments. Additionally, we perform an ablation study that evaluates each major architectural component. Finally, we qualitatively explore the learned embedding space, and show that semantic relationships are discovered across modalities.

UR - http://www.scopus.com/inward/record.url?scp=85075426930&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075426930&partnerID=8YFLogxK

U2 - 10.1145/3357384.3358001

DO - 10.1145/3357384.3358001

M3 - Conference contribution

AN - SCOPUS:85075426930

T3 - International Conference on Information and Knowledge Management, Proceedings

SP - 1993

EP - 2002

BT - CIKM 2019 - Proceedings of the 28th ACM International Conference on Information and Knowledge Management

PB - Association for Computing Machinery

ER -

Jenkins P, Farag A, Wang S, Li Z. Unsupervised representation learning of spatial data via multimodal embedding. In CIKM 2019 - Proceedings of the 28th ACM International Conference on Information and Knowledge Management. Association for Computing Machinery. 2019. p. 1993-2002. (International Conference on Information and Knowledge Management, Proceedings). https://doi.org/10.1145/3357384.3358001