International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Contents lists available at ScienceDirect

International Journal of Applied Earth

Observation and Geoinformation
journal homepage: www.elsevier.com/locate/jag

Mapping the landscape and roadmap of geospatial artificial intelligence

(GeoAI) in quantitative human geography: An extensive systematic review
Siqin Wang a, b, c, *, Xiao Huang d, *, Pengyuan Liu e, Mengxi Zhang f, Filip Biljecki g, h, Tao Hu i,
Xiaokang Fu j, k, Lingbo Liu j, Xintao Liu l, Ruomei Wang b, Yuanyuan Huang b, Jingjing Yan v,
Jinghan Jiang m, Michaelmary Chukwu v, Seyed Reza Naghedi v, Moein Hemmati n,
Yaxiong Shao o, Nan Jia p, Zhiyang Xiao l, Tian Tian q, Yaxin Hu r, Lixiaona Yu i, Winston Yap g,
Edgardo Macatulad g, Zhuo Chen s, Yunhe Cui t, Koichi Ito g, Mengbi Ye g, Zicheng Fan g,
Binyu Lei g, Shuming Bao u
a
Spatial Sciences Institute, University of Southern California, Los Angeles, United States
b
University of Queensland, Brisbane, Australia
c
School of Science, RMIT University, Melbourne, Australia
d
Department of Environmental Sciences, Emory University, Atlanta, United States
e
Future Cities Lab Global, Singapore-ETH Centre, Singapore
f
Carilion School of Medicine, Virginia Tech, Blacksburg, VA, United States
g
Department of Architecture, National University of Singapore, Singapore
h
Department of Real Estate, National University of Singapore, Singapore
i
Department of Geography, Oklahoma State University, Stillwater, OK, United States
j
Centre for Geographic Analysis, Harvard University, Cambridge, MA, United States
k
State Key Laboratory of Information Engineering in Surveying, Mapping, and Remote Sensing, Wuhan University, Wuhan, China
l
Department of Land Surveying and Geo-informatics, The Hong Kong Polytechnic University, Hong Kong, China
m
Department of Environmental Science Policy and Management, University of California Berkeley, Califorina, United States
n
Environmental Dynamics Program, Graduate School and International Education, University of Arkansas, Fayetteville, AR, United States
o
Department of Earth Atmosphere and Environment, Northern Illinois University, Dekalb, IL, United States
p
Center for Systems Integration and Sustainability, Department of Fisheries and Wildlife, Michigan State University, Michigan, United States
q
Department of Urban Planning, School of Urban Design, Wuhan University, Wuhan, China
r
The Fu Foundation School of Engineering and Applied Science, Columbia University, New York, NY, United States
s
Harrington Heart and Vascular Institute, University Hospitals, and School of Medicine, Case Western Reserve University, Cleveland, OH, United States
t
Department of Geography, University of Connecticut, Storrs, CT, United States
u
China Data Institute, Michigan, United States
v
Department of Geosciences, University of Arkansas, AR, United States

A R T I C L E I N F O A B S T R A C T

Keywords: This paper brings a comprehensive systematic review of the application of geospatial artificial intelligence
Geospatial artificial intelligence (GeoAI) in quantitative human geography studies, including the subdomains of cultural, economic, political,
GeoAI historical, urban, population, social, health, rural, regional, tourism, behavioural, environmental and transport
Human geography
geography. In this extensive review, we obtain 14,537 papers from the Web of Science in the relevant fields and
Geographic subdomains
Systematic review
select 1516 papers that we identify as human geography studies using GeoAI via human scanning conducted by
several research groups around the world. We outline the GeoAI applications in human geography by system­
atically summarising the number of publications over the years, empirical studies across countries, the categories
of data sources used in GeoAI applications, and their modelling tasks across different subdomains. We find out
that existing human geography studies have limited capacity to monitor complex human behaviour and examine
the non-linear relationship between human behaviour and its potential drivers—such limits can be overcome by
GeoAI models with the capacity to handle complexity. We elaborate on the current progress and status of GeoAI
applications within each subdomain of human geography, point out the issues and challenges, as well as propose

* Corresponding authors at: Spatial Sciences Institute, University of Southern California, Los Angeles, United States (S.Wang).
E-mail addresses: [email protected] (S. Wang), [email protected] (X. Huang).

https://doi.org/10.1016/j.jag.2024.103734
Received 7 December 2023; Received in revised form 26 January 2024; Accepted 23 February 2024
Available online 11 March 2024
1569-8432/© 2024 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

the directions and research opportunities for using GeoAI in future human geography studies in the context of
sustainable and open science, generative AI, and quantum revolution.

1. Introduction multifaceted and non-linear (Li, 2022).

Human geography is the branch of geography that studies spatial
Geospatial artificial intelligence (GeoAI) is an emerging and prom­ relationships between human communities, cultures, economies, and
ising research field that integrates AI with geospatial science to resolve their interactions with the environment (Hoggart, 2002). Whereas
problems and issues of geographic nature (Li and Hsu, 2022). The physical geography concentrates on spatial and environmental processes
development of GeoAI brings the advantages of traditional AI studies in that shape the natural world and tends to draw on the natural and
computer science to geographic research by empowering its quantitative physical sciences for its scientific underpinnings and methods of inves­
methods with revolutionary technologies including machine and deep tigation, human geography concentrates on the spatial organization and
learning, high-performance computing power, and big data mining(Liu processes shaping the lives and activities of people, and their in­
and Biljecki, 2022). Such emerging AI-oriented research tendency is teractions with places and nature (Gregory et al., 2011). Human geog­
particularly important for geographic studies in the era of big data, raphy consists of a number of sub-disciplinary domains that focus on
given more than 80 % of big data contain spatial information (Leszc­ different elements of human activity and organization (Gregory et al.,
zynski and Crampton, 2016). As the advocacy of the Tobler’s first law of 2011), mainly including (Fig. 1) cultural, economic, political, historical,
geography—“everything is related to everything else, but near things urban, population, social, health, rural, regional, tourism, behavioural,
are more related than distant things” (Miller, 2004), GeoAI enables re­ environmental and transport geography. What distinguishes human ge­
searchers better monitor human behaviours and the surrounding envi­ ography from other related disciplines, such as development, eco­
ronment which are often spatially dependent and autocorrelated. The nomics, politics, and sociology, are the application of a set of core
recent breakthrough in GeoAI and more specifically deep learning fa­ geographical concepts and notions that the world operates spatially and
cilitates the growth of a new research paradigm integrating data science temporally, and that social relations were thoroughly grounded in and
and geography to analyse, mine, and visualise large volumes of spatio­ through of place and environment—where the implementation of GeoAI
temporal data, as well as enables researchers better capture the human- is well positioned and urgently needed. Although a range of review
environment relationship given such relationship is complex, papers give attention to various related topics such as the application of

Fig. 1. Conceptual framework to shape the review scope.

2
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

deep learning in geography (Li and Hsu, 2022), GeoAI applications in of human geography—which our study fulfils.
urban planning and development (Alastal and Shaqfa, 2022) and urban We conduct a systematic review on the implementation of GeoAI in
geography (Liu and Biljecki, 2022), GeoAI approaches for complex quantitative human geography including the subdomains listed above to
geomatics data (Pierdicca and Paolanti, 2022), unsupervised machine 1) provide a holistic picture of the state-of-the-art GeoAI techniques and
learning in urban studies (Wang and Biljecki, 2022), and more broadly applications that have been used in human geography as well as data
GeoAI in social science from a scoping review perspective (Li, 2022), sources that were used to support GeoAI; 2) outline the future directions
what is lacking from the current scholarship is a holistic, comprehensive, for geographers to grasp the AI-oriented opportunities whilst at the same
and systematic understanding of GeoAI application in various domains time the future challenges and risks that need us to think critically and

Fig. 2. PRISMA workflow for paper selection.

3
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

tackle specifically. Failing to do so could be costly and left behind in the Table 1
mainstream of science as others leverage insights from the growing data Inclusion and exclusion criteria used in the paper selection through human
deluge. scanning.
Essential inclusion criteria: Essential exclusion criteria:
2. Review method human-centred; related to human/ non-human-centred; focusing on
society, human AI or human physical aspects; or techniques centred;
intelligence; focusing on the impact of without focusing on their impact on
2.1. Data retrieval other phenomena on human society human society and population.
and population.
We employed the standard systematic review methodology (Moher Exemplary papers are: Exemplary papers are:
• Quantitative studies (including data, • Qualitative summary; outline of study
et al., 2010), known as the Preferred Reporting Items for Systematic
methods, modelling tasks, and results) progress; review papers
Reviews and Meta-Analyses (PRISMA) method (Fig. 2), to collect, scan • To predict or estimate demographic • Physical aspects of urban, and
and select appropriate papers within our research scope (detailed in and socioeconomic aspects of neighbourhood features, land use and
Section 2.2). Following the existing review work (e.g.,Li and Hsu, 2022; populations change
Liu and Biljecki, 2022; Wang and Biljecki, 2022), we obtained papers • Social sensing via social media to • Techniques comparisons;
study human behaviour (e.g., methodological oriented (e.g., IoT, 5G,
relevant to GeoAI application in human geography from the Web of perception, attitude, opinions, mental technical resolution, software, etc)
Science (WOS), as one of the most popular academic databases, based on signals, natural languages, and
several sets of predefined search syntax: 1) paper topics (in the WOS semantics towards some phenomena)
advanced searching engine, these terms were searched through topics • Public health • Climate change modelling without
considering effects on human and
which include article titles, abstracts, and keywords) relevant to GeoAI,
society
including “artificial intelligence”, “geospatial artificial intelligence”, • Economic activities (e.g., housing • Remote sensing only (e.g., object
“AI”, “GeoAI”, “machine learning”, “deep learning” and “neural prices, settlement, etc) and sharing detection, data fusion, etc)
network”; 2) paper topics relevant to human geography, including economy (e.g., sharing cars, etc)
“urban*” (the asterisk expands the search to include variations of the • Political activities (e.g., agents; • Weather and air pollution prediction
stakeholders; political economy; without impacts on human
key syntax, such as urbanization), “city*”, “place*”, “human*”, “geo*”,
legitimation)
as well as the name of 14 subdomains shown in Fig. 1. Their intersection • Urban governance and management; • Survey and workshop to educate
signifies the focus of our review; 3) article type as “peer-reviewed human centred urban planning people to study AI
journal articles” with the language as “English”; 4) disciplines as • Social phenomenon (e.g., migration, • Transport modelling without impacts
gentrification, etc) on human (e.g., system performance,
“environmental studies”, “environmental sciences”, “geography phys­
etc)
ical”, “geography”, “green sustainable”, “science technology”, “engi­ • Transportation choices and
neering civil”, “urban studies”, “regional & urban planning”, behaviours (e.g., transport sharing,
“sociology”, “social sciences”, “interdisciplinary”, “social sciences”, walkability, etc)
“mathematical methods”, “humanities”, and “multidisciplinary” to • Human-eye measures (e.g., visual
quality, street view, visual greenness,
further refine the research; 5) publication timespan as “in and after
perceived measures, etc)
2010) given AI techniques started to rapidly develop after 2010 with the
rise of social media, smartphones, and the internet of things (Van Roy
et al., 2020). 3. Statistical outline of GeoAI applications in human geography
With all these settings, the initial search results display 45,350 pa­
pers and the majority of such papers were relevant to remote sensing and This section summarises how GeoAI applications have been applied
physical geography without focusing on their impact on human society in Human Geography studies across different subdomains in terms of
and population as well as a proportion of these papers fell into other modelling tasks, study areas and data types by using a Sankey diagram
disciplines (e.g., food industry and hydrology) beyond our review scope. (Fig. 3). The full version containing all categories is provided in Sup­
Accordingly, we added one more set of search syntax—“remote plementary Figure S1. This visualisation shows that 1) the top three
sensing”—as the exclusion criteria to exclude papers, that are not subdomains using GeoAI in human geography include urban geography,
human/population centred. Finally, the research results by the time of transport geography and environmental geography (more details pro­
this study, February 1st, 2023, show that 14,537 papers were vided in Fig. 4); 2) the top three study areas where GeoAI was applied
obtained—a substantial number for a systematic review, but a reason­ most frequently include China, US and India (more details provided in
able one considering the broad scope of the paper and one that can be Fig. 5); 3) the top three data types that were most widely used in GeoAI
manageable for human scanning in a large team. The paper list was include remote sensing imageries, street view imageries and transport
downloaded with attributes including the publication year, author network (more details provided in Fig. 6); 4) the top three modelling
name, article title, journal, keywords, abstract, and the number of ci­ tasks implemented by GeoAI include predication (i.e., the non-linear and
tations by the time of this study. linear regression) and image-based classification (more details provided
in Fig. 7).
It is evident from Fig. 4 that the usage of GeoAI in human geography
2.2. Scanning papers with human efforts has significantly increased since 2018. In 2019, there were 168 papers,
which rose to 526 by 2022. Out of the 1512 selected papers, the majority
It was imperative to select the pre-obtained papers specifically cen­ of them, accounting for 29.1 % (440), were related to urban geography.
tred around human society, population, and their perception and rela­ Transport geography followed with 16.4 % (248), environmental ge­
tionship with the surrounding environment—as the essential notion of ography with 14.1 % (213), and health geography with 10.6 %. Inter­
human geography that the spatial organization and processes where the estingly, the number of papers published in health geography showed a
lives and activities of population immerse are shaping their interactions sharp increase from 2020 to 2021, which could be attributed to the
with places and nature. Thus, we conducted the paper scanning with COVID-19 pandemic outbreak.
human efforts from March 1st to April 1st, 2023 based on the inclusion GeoAI has been applied in various study areas, with China leading at
and exclusion criteria (Table 1). Essentially, papers need to be included 33.3 % (5 2 9) followed by the US at 17.6 % (2 8 0), India at 4 % (63),
if they are quantitative studies in any subdomain of Human Geography, multiple countries worldwide at 3.8 % (60), and UK at 3 % (48) (Fig. 5).
as well as being human-, population- and society-centred within the The subdomain of urban geography is the most popular among all study
application of GeoAI; otherwise, they should be excluded.

4
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Fig. 3. Sankey diagram showing the distribution of study areas, data types and modelling tasks GeoAI achieved in each subdomain of human geography (top 15
categories included); the full version containing all categories is provided in Supplementary Fig. S1.

areas where GeoAI has been applied. For example, 36.9 % of 529 papers varying across subdomains.
in China and 23.6 % of 280 papers in the US are in the domain of urban The modelling tasks that have been implemented by GeoAI appli­
geography. The second and third most popular study areas where GeoAI cations in Human Geography studies are classified into broad and sec­
has been applied are transport geography and environmental geogra­ ondary categories (Fig. 7) including 1) classification (e.g., image-based,
phy, respectively. vector-based, sequential-data-based, number-based, and text-based); 2)
The types of data that have been used in GeoAI applications in prediction (e.g., linear and non-linear regression); 3) simulation (e.g., at
Human Geography are classified into broad and secondary categories the aggregated and individual level); 4) embedding (i.e., the reduction
(Fig. 6) including 1) raster-based imageries (e.g., remote sensing images, of data dimensions; feature extraction); and 5) geolocating (i.e., geo­
street view images, and geotagged photos); 2) vector-based spatial data parsing). More details of the modelling tasks are provided in Table 3. We
(e.g., points of interest, building, built environment, land use and land can observe some common patterns of GeoAI modelling tasks across
cover maps, and 3D point clouds); 3) vector-based spatiotemporal flow various subdomains of human geography—prediction (particularly non-
data (e.g., transport network, call records, mobile phone signals, linear regression) and classification (particularly image-based classifi­
smartphone apps, credit card transactions and traffic records); 4) official cation) have been widely employed in different subdomains of human
data provided by government or other sectors (e.g., census, geography, regardless of the number of published papers. GeoAI in
government/authority-lead survey, health, economic data, tourists, urban geography also conducted a range of simulation tasks, in partic­
crime, meteorological data, and other official statistics); 5) sound, video ular the aggregated level simulation (e.g., using cellular automata, deep
and texts; and 6) self-conducted survey. More details of the data clas­ neural network, and deep enforcement learning models).
sifications are provided in Table 2. The data types used in GeoAI are

5
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Fig. 4. (A) The number of papers published over the years and in each subdomain by accounting for the primary subdomain (the statistics accounting for both the
primary and secondary subdomains are provided in Supplementary Table S1. The less common subdomains are plotted separately due to their minor share. The year
2023 is included in the review but not in this plot since it was not complete at the time of the writing.; (B) the total number of papers published in each subdomain
from 2010 to 2023.

Fig. 5. A number of papers with empirical studies focusing on different regions, and the share of their categories. Note: there are a total of 100 countries studied in
our selected papers; this bar chart only includes the top score of regions most frequently studied and the full list of them is provided in Supplementary Table S2.

4. How GeoAI enhances each subdomain of human geography 4.1. Urban geography

This section systematically and comprehensively summarises how GeoAI plays a pivotal role in Urban Geography, analyzing urbani­
GeoAI has been applied to 13 subdomains of Human Geography and zation, urban changes, and hazards. Advancements in GeoAI include
advanced the current studies in terms of analytical approaches, integrating CA with machine learning methods ((e.g.,Azari et al., 2016;
emerging data, and research scopes. Kafy et al., 2021; Rienow and Goetzke, 2015; Zhang and Xia, 2022;
Zhang, Liu, et al., 2019) and neural networks (He et al., 2018; Ullah
et al., 2019; Yang et al., 2019), using varied data sources like satellite

6
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Fig. 6. Data types are classified into the broad and secondary category.

Fig. 7. Modelling tasks are classified into the broad and secondary categories.

images (Qian et al., 2020; Wu et al., 2010) and census data (Feng et al., analysis. As evidenced by the paper collection, thanks to the rapid
2019; Wu et al., 2010) to support precise and high-resolution simula­ enrichment of multiple data sources, we have witnessed a clear trend in
tions in the cities. Beyond CA, other methods, such as CNNs (e.g., Chen exploring deep learning and its combination of machine learning and
et al., 2020b) and generative models (e.g., Ibrahim et al., 2021), also statistical methods in studying urban issues and discovering new pat­
show their potential to model the spatial and spatiotemporal in­ terns of cities that facilitate decision-making or assist downstream
teractions of urban objects in the phases of urbanisation. To study urban analytics.
places and functions, with the development of smart urban in­
frastructures (e.g., smart cards for transportation), human-sensing data
(e.g., geo-tagged text and images) and digital platforms (e.g., Open­ 4.2. Transport geography
StreetMap, Mapillary), intelligent computational models, such as natu­
ral language processing and computer vision-based (e.g., CNNs) The application of GeoAI in transport geography is a rapidly evolving
analytical frameworks have massively contributed to the understanding field that has gained significant attention in recent years. The use of
of urban places (e.g., Andrade et al., 2020; Kim, 2019; Luo et al., 2022a; machine learning and AI techniques has the potential to improve
Wu et al., 2023; Zhang et al., 2023; Zhu et al., 2020) and supporting transportation systems while providing insights into human travel
urban functional zones planning (e.g., Jiang et al., 2015; Zhai et al., behaviour. To date, numerous articles have focused on specific modes of
2019; Zhang et al., 2019a). Urban hazards cover a wide range of topics, transportation such as urban rail transit (e.g., Wang et al., 2021a; Zhang
such as flooding, heat waves, and fire emergencies. Tools developed et al., 2020b) bike-sharing (e.g., Fontes et al., 2022; Yang et al., 2018),
based on CNNs with bird’s eye-level remote sensing and human eye- and ride-hailing services (e.g., Huang et al., 2021; Niu et al., 2019),
level street view images (e.g., Bao et al., 2019; Feng and Sester, 2018; while others have addressed broader topics such as traffic forecasting (e.
Li et al., 2022a) are adopted to support hazard detection and urban g., Zhang et al., 2019c; Zhang et al., 2020f), crash prediction (e.g., Hu
reconstructions. With the help of smart devices and social media plat­ et al., 2022; Wei et al., 2022), and mobility pattern recognition (e.g.,
forms, natural language processing methods, such as geo-referencing (e. Heredia et al., 2021; Lv et al., 2021). Of particular interest is the growing
g., Zhang et al., 2021) and content analysis (e.g., Agonafir et al., 2022), trend in exploring the use of deep learning models for analysing images
are exclusively developed to support hazards localisation and severity (e.g., street views and user-generated pictures), and other sensor data (e.
g., traffic sensors, GPS pins, and environmental sensors), to understand

7
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Table 2 Table 3
Type of data that have been used in GeoAI related Human Geography studies. Modelling tasks and methods that have been implemented in GeoAI related
Broad category Secondary Exemplary data sources
Human Geography studies.
category Broad Secondary Exemplary models and algorithms
category category
Imageries (raster) Remote sensing Land use and land cover raster
imageries maps, night-time data, LIDAR, Classification Image-based • Convolutional Neural Networks (CNNs)
MODIS (LeNet, AlexNet, ResNet, VGG, Inception,
Street view User/vehicle-generated imageries, EfficientNet, DenseNet, MobileNet)
imageries Google/Baidu street view imageries • Artificial Neural Network
Geotagged photos Geotagged (social media) photos • Support vector machine (SVM) and one-
such as Flickr class SVM
Spatial data (vector) Points of Interest Location of public facilities, parks, • Gradient Boosting algorithms (GBM) (e.
and bus stops g., XGBoost, LightGBM, Catboost)
Building Cadastral maps/building footprints • K-Nearest Neighbour
Built Measures of road density, distance • Naïve Bayes Algorithm
environment to the nearest stations • Deep Belief Networks (DBNs)
Land use and land Polygon layers of land use and land • Autoencoder (AE)
cover cover • Siamese networks
3D point clouds 3D point clouds data including • Isolation forest
vertical dimension • Local outlier factor
Spatial-temporal flow Transport Mobility network, road network, • Angle-based outlier detector
data (vector) network smart card-generated data, flight • Histogram-based outlier detection
tickets • Autoencoders (variational types)
Call records 911 hotline calls, municipal • Hidden Markov models
complaint records • Fuzzy logic-based outlier detection
Mobile phone Safegraph • Deep-learning based methods
signal (Conditional neural network, RNN)
Smartphone app Social media app/website- • YoLo model family
generated data (e.g., Baidu/ • R-CNN model family (R-CNN, Fast R-
Google/Facebook mobility data), CNN, Mask R-CNN, R-FCN, Cascade R-
tweets/geotagged tweets, online CNN)
bike-sharing information • CenterNet model family (Single Shot
Credit card Credit card records with Detector (SSD), DSSD, RON, CornerNet)
transactions spatial–temporal information • Histogram of Oriented Gradients (HOG)
Traffic Traffic crashes, bike usage, accident • Region-Based Segmentation
records, and emergency events. • Edge Segmentation
Official data provided Census Census data provided by the • K-Means
by gov/organisation/ government • Convolutional Encoder-Decoder Archi­
authority Gov/authority- Disaster evacuation data in the US, tecture (e.g. SegNet, U-Net, Fully Con­
lead survey HILDA data in Australia volutional Networks (FCN)
Health Clinic/patient data, healthcare • Multi-Scale and Pyramid Network Based
facilities Models (FPN)
Economic Housing, insurance, sale and retail • Pyramid Scene Parsing Network
records, tax assessment, residential (PSPNet), Mask R-CNN, Fast R-CNN)
settlement • Dilated Convolutional Models and
Tourist Tourist statistics DeepLab Family
Crime Crime records Vector-based • Spatially constrained multivariate
Meteorological Weather station-generated data (e. clustering
g., wind, rainfall, humidity) • Multivariate clustering
Other statistics Bio environment (e.g., species) • Density-based clustering
Sound, video and texts Sound Voice, sound records • Image segmentation
Video Traffic video • Hot spot analysis
Text content Texts without spatiotemporal • Cluster and outlier analysis
information • Space-time pattern mining
Photo Static photos without • Hierarchical clustering analysis (HCA)
spatiotemporal information • Density-based spatial clustering of
Self-conducted survey — Primary data collected via online/ applications with noise (DBSCAN)
offline interviews, questionnaires • Spectral clustering
• Affinity propagation (AP)
• Gaussian mixture model (GMM)
the built environment’s impact on travel behaviour (e.g., Li et al., 2023a; Sequential (time) • Hidden Markov Models
Li et al., 2023b; Liu et al., 2022). Additionally, researchers are data-based • Long Short-Term Memory networks
(LSTM)
increasingly interested in the application of intelligent algorithms (e.g.,
• Recurrent Neural Networks (RNN)
tree-structure models, graph neural networks, and recurrent neural • Conditional Random Fields
networks) for predicting travel patterns, including mode choice (e.g., • Distance-based (e.g., KNN, dynamic
Zhang et al., 2020a), passenger flow (e.g., Zhang et al., 2020a), and warping)
traffic volume (Zheng et al., 2021). To support transportation planning • Interval-based (e.g., time-series forest)
• Dictionary-based (e.g., Bag of SFA
efforts, various tools have been developed, such as a planning support Symbols (BOSS))
tool for street network design (Fang et al., 2022). Moreover, there is a • Frequency-based (Random Interval
surge of research activities proposing novel approaches to Specteal Ensemble (RISE))
transportation-related challenges, such as detecting traffic incidents • Shapelet-based (e.g., shapelet transform)
Number (non- • Support vector machine (SVM)
using social media data (e.g., Chang et al., 2022), identifying critical
spatial)-based • Gradient Boosting algorithms (GBM) (e.
transfer zones for coordinating transit (e.g., Qiu et al., 2022), and pre­ g., XGBoost, lightGBM, CatBoost)
dicting the demand for bike-sharing and ride-hailing services (e.g., Chen • Decision tree / Random Forest
et al., 2020a; Zhang and Zhao, 2022). These approaches have the (continued on next page)

8
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Table 3 (continued ) potential to transform the way transportation systems are planned,
Broad Secondary Exemplary models and algorithms designed, and managed. Overall, we observe that GeoAI has emerged as
category category a highly promising avenue for transforming the transportation industry.
• Means algorithm
Nevertheless, as many scholars have pointed out, the potential chal­
• Fuzzy logic-based algorithms lenges that must be overcome include integrating disparate data sources,
• DBSCAN addressing privacy concerns, and devising efficient algorithms that can
• Spectral clustering process immense quantities of transportation data in real time. None­
Hierarchical clustering
•
theless, the commendable progress achieved thus far has laid the
• Affinity Propagation
Text-based • Latent Dirichlet allocation (LDA) / RNN groundwork for sustained research in the GeoAI domain of Transport
• Word2Vec Geography, thereby facilitating continued improvements in trans­
• Doc2Vec portation systems.
• Bag-of-words model
• n-gram model
• Transformers-based methods (BERT,
4.3. Environmental geography
XLM, GPT, RoBERTa, XLNet, DistilBERT
etc) Environmental geography has gained significant importance in
• ELMo recent years due to pressing global environmental challenges, such as
RNN
climate change, population growth and urbanization, environmental
•
• LSTM
• Word2Vec pollution, and an increasing number of natural disasters. By leveraging
• Doc2Vec the power of advanced machine learning and artificial intelligence
• Bag-of-words model techniques, GeoAI enables the processing and analysis of large and
• n-gram model complex environmental data with the integration of diverse data sources
• Transformers-based methods (BERT,
XLM, GPT, RoBERTa, XLNet, DistilBERT
(e.g., Ning et al., 2020), remote sensing imagery, street view images (Wu
etc) et al., 2020), social media posts (Ning et al., 2020), and census, real-time
• ELMo monitoring, and enhanced predictive capabilities. For example, GeoAI
Prediction Linear • Generalized linear model (GLM), techniques have been employed for land use and land cover changes
including Lasso regression, Ridge
detection and prediction (e.g., Apollonio et al., 2016; Handayanto et al.,
regression, Polynomial Regression,
Bayesian linear regression; Logistic 2017), climate change modelling and forecasting (e.g., Logan et al.,
regression, Gamma regression, Poisson 2020; Shen et al., 2020), disaster management and risk assessment (e.g.,
regression, Bernoulli regression, Binomial Bui et al., 2019; Feng and Sester, 2018; Talukdar et al., 2020), water
regression, Multinomial regression, resource management (e.g., Obringer and White, 2023; Zhang et al.,
Exponential regression, (Inverse)
Gaussian regression
2018), environmental pollution monitoring (e.g., Acheampong and
Non-linear • Support vector machine Boateng, 2019; Cole et al., 2020), and urban planning and sustainable
• Artificially Neural Network (ANN) development (e.g., Richards and Tunçer, 2018; Wu et al., 2020). As
• Gradient Boosting algorithms (GBM) (e. GeoAI continues to evolve, it holds the potential to significantly improve
g., XGBoost, lightGBM, CatBoost)
our understanding of environmental geography and provide essential
• Empirical Bayesian Kriging regression
prediction insights for effective decision-making to address pressing environmental
• Forest based prediction (random forest, issues.
decision tree)
• Graph Convolutional Neural Network 4.4. Health geography
• Generalised additive model (GAM) and
GeoGAM
• Bayesian hierarchical model (BHM) The implementation of GeoAI can be easily identified in many
• Second-dimension spatial association health-related areas, including infectious epidemiology, social media
• Geographically optimal similarity model analysis, built environment, and environmental epidemiology. GeoAI
Simulation Aggregated level Cellular Automata
has been used to identify and forecast the spread of infectious diseases
•
(areal-based) • Deep neural network
• Deep enforcement learning (Guo et al., 2017), but the booming of this application was not until the
• Tabular Q-learning hit of the COVID-19 pandemic that various models have been developed
Individual level • Agent based modelling and used for COVID-19 prediction (Ghahramani and Pilla, 2021; Guo
(agent-based) and He, 2021; Tomar and Gupta, 2020), especially at the early stage of
Embedding — • Principal component analysis (PCA)
• Independent Component Analysis (ICA)
the pandemic. Other application related to COVID-19 includes
• Linear Discriminant Analysis (LDA) measuring the association between mitigation policies and COVID-19
• Locally Linear Embedding (LLE) transmission (Zhang et al., 2020c), maximizing the assessed popula­
• t-distributed Stochastic Neighbour tion in the shortest possible time for mobile assessment agents (Simsek
Embedding (t-SNE)
and Kantarci, 2020), and monitoring social distancing using video se­
• Auto-encoder model family
• enerative adversarial network (GAN) quences (Ahmed et al., 2021). GeoAI has also been applied to analyse
• Isomap social media data to help with the early detection of the distribution of
• Hessian Eigenmapping COVID-19 (Golder et al., 2020). It has also been widely used to under­
• Spectral embedding stand the distribution and public opinion of other diseases, including
Multi-dimensional Scaling (MDS)
estimating obesity prevalence (Cesare et al., 2019), understanding
•
• Kernel PCA
• Graph Neural Networks (GraphSAGE, autism-related antivaccine beliefs (Tomeny et al., 2017), monitoring the
GCN) well-being of transit riders (Tran et al., 2023), and measuring urban-
Geolocating — • Geoparsing models regional disparities of mental health signals (Wang et al., 2022b).
GeoAi was mainly used to identify tweets and to measure emotions
within a body of text. There has been a long-term interest in applying
GeoAI to better evaluate built environment (e.g., cities, buildings, and
greenspace) (Larkin and Hystad, 2019) and measure its association with
other health outcomes, including COVID-19 transmission (Kawlra and

9
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Sakamoto, 2021), mental health status (Helbich et al., 2021; Wang et al., largely categorised into the subdomain of health geography.
2019a), the well-being of elderly (Wang et al., 2019b), and pedestrian
emotion (Xiang et al., 2021). Most of this work applies GeoAI to identify 4.7. Social geography
the features of the built environment through analysing street view
images. GeoAI has also been advancing modelling technics in environ­ Studies in social geography are largely mingling with other sub­
mental epidemiology in accessing air pollution and measuring its impact domains of human geography, including behavioural geography, urban
on health. Recent applications include predicting PM2.5 concentration geography and environmental geography. GeoAI empowers the methods
during the pandemic (Yang et al., 2022b), generating daily surface that were used in the traditionally sociological/urban studies to examine
concentration maps for PM2.5 and O-3 during a wildfire (Reid et al., the relationship between urban built/natural environment and the
2019), and predicting air quality (Xiao et al., 2021). It also helps to observed social phenomenon, including gentrification and community
model the association between air pollution exposure and early cogni­ deprivation (Alizadeh et al., 2018; Li et al., 2019b; Reades et al., 2019;
tive skills among children (Stingone et al., 2017). Talukdar et al., 2021); to evaluate the impacts of social events via large-
scale spatiotemporal crowdsourcing data (e.g., geo-tagged social media
4.5. Economic geography data (Zhu et al., 2018); to address some unique social problems such as
modern slavery (Lavelle-Hill et al., 2021); to visualise the urban social
The GeoAI applications in economic geography have largely crossed change of neighbourhoods (Lee and Rinner, 2015); to identify poverty
over urban geography, in particular, land use and land change (e.g., Hu and urban slums (Yin et al., 2020) and explore the potential driving
et al., 2016a; Levers et al., 2018), given land provides space for eco­ factors of poverty (Luo et al., 2022b); to quantify neighbourhood mixing
nomic activities and relevant costs occurring along the process of land (Hipp et al., 2017); to evaluate social justice (Debnath et al., 2021),
changes link to economic development and evaluation. GeoAI models, housing inequality (Knaap, 2017), and social inequality in the selection
especially non-linear models, have been widely employed to predict and of transport modes (Zhou et al., 2019); and to monitor social wellbeing
estimate sale prices of real estate properties (Čeh et al., 2018; Rafiei and of different social groups (Brown et al., 2021). GeoAI also has implica­
Adeli, 2016) and income levels (Ivan et al., 2020); to generate a wide tions for the management of social organisations and public policy via
range of socioeconomic indicators based on other data sources (e.g., examining social network (Choudhury et al., 2022) and organizational
remote sensing imageries, open street maps and social media data) geosocial network (Zhao et al., 2022), as well as to facilitate citizen
(Feldmeyer et al., 2020; Roumiani and Mofidi, 2022); to explore the engagement in urban governance (Siyam et al., 2020) and to predict the
potential driving factors for housing and rental prices(Ma et al., 2020; socioeconomic status of urban neighbourhoods (Diou et al., 2018).
Yoo et al., 2012) or economic consumptions and activities across Similar to some studies in behavioural geography, GeoAI has been also
different industries (e.g., electric appliances, retail industry)(AL- used to track and predict neighbourhood crime (Amiruzzaman et al.,
Musaylh et al., 2021; Rao and Ummel, 2017); or to evaluate the rela­ 2021) and examine how it links to neighbourhood visual appearance
tionship between management and organizational performance (Lin (Reier Forradellas et al., 2020) and the socioeconomic attributes of lo­
et al., 2022). Those relationships among economic activities and their cations (Doi et al., 2021). Lastly, GeoAI helps to generate new datasets
potential factors are complex, non-linear and mingling with various from the perspective of social science, including the social vulnerability
confounders, where GeoAI provides advanced approaches to generalise index (Alizadeh et al., 2018) and livelihood vulnerability index
such interrelationships. In addition, GeoAI helps to develop planning (Talukdar et al., 2021), which can be further used in cross-disciplinary
and decision support system (Demetriou et al., 2012), urban renewal studies.
policy analysis based on a wide range of crowdsourcing data (Auerbach
et al., 2020) or to simulate housing rentals (Chen et al., 2016) with the 4.8. Tourism geography
advantages of visualising its spatiotemporal patterns—these end-user
applications and policy implications can readily benefit government, The application of Geo AI in tourism geography has significantly
public sectors and authorities on economic planning and policy making. advanced the field by providing insights into various aspects of tourism
dynamics and management. Key areas of focus include analysing tourist
4.6. Behavioural geography behaviour and movement patterns (e.g., Chang et al., 2019; Zhang et al.,
2019b), understanding visitor experiences (e.g., Song et al., 2021; Zhang
One cutting-edge advance of GeoAI is to quantify the behaviour of et al., 2020d), building an attraction evaluation and recommendation
human beings which could be difficult to achieve by other approaches. system (Giglio et al., 2019; Sun et al., 2015), evaluating sustainability (e.
Such human behaviours include human perception (e.g., the sense of g., D’Uva and Rolando, 2023; Roumiani et al., 2023), assessing the
safety, anxiety, depression, beauty, and happiness) to the visual features quality of public spaces (e.g., Ghahramani et al., 2021a; Li et al., 2022b),
of the built environment (e.g., green and blue space) and natural envi­ and monitoring environmental impacts (e.g., Furgała-Selezniow et al.,
ronment (Choi et al., 2022; Ramírez et al., 2021; Rossetti et al., 2019; 2021). These studies use various data sources including surveys (such as
Zhang et al., 2018)as well as certain social phenomena, events and crises Hou et al., 2021), geotagged photos (such as Zhang et al., 2019b), points
(e.g., COVID-19; disease treatment-seeking behaviours) (Kim et al., of interest (such as Zhang et al., 2020d), smartphone apps and signals
2022). GeoAI has also been widely applied to explore the relationship (such as Crivellari and Beinat, 2020; Song et al., 2021), transport net­
between environment and human decisions on travel (e.g., active works (such as Nuzzolo and Comi, 2016), remote sensing imagery (such
commuting, walkability, bikeability, and transport demand) (Ding et al., as Sun et al., 2021), street view images (such as Kruse et al., 2021), and
2018; Ki and Lee, 2021; Molina-García et al., 2019; Rossi et al., 2019; text content from social media (such as Zhang et al., 2020d). Accord­
Tran et al., 2020), linguistic and semantic expressions (Wirz et al., ingly, the GeoAI methods used in tourism geography mainly include
2018), spatiotemporal trajectory (Torrens et al., 2011), and consumers’ text-based techniques like topic modelling (van Weerdenburg et al.,
behaviours (Dias et al., 2021): such human-environment relationships 2019), and sentiment analysis (Ghahramani et al., 2021b), computer
are confounding, complex which can be rarely captured by linear vision models such as deep learning models (Kang et al., 2021), other
models. The prediction of crime (criminal rates, locations and patterns) machine learning techniques like random forest (Li et al., 2019a) and
and social lifestyles (Ben Zion and Lerner, 2018; Kadar and Pletikosa, XGBoost (Kang et al., 2022), as well as spatial clustering models like
2018) is another stream of GeoAI applications in behavioural geogra­ density-based clustering method (DBSCAN) (Sun et al., 2015). Overall,
phy. Since the outbreak of COVID-19, GeoAI become popular to capture GeoAI has been widely used to provide insights for policy decisions,
people’s altitude, opinion, mental reaction and connection to the global improve tourism management, and contribute to a better understanding
pandemic (e.g., Wang et al., 2022b), although such studies have been of human behaviour and perception of urban spaces and travel

10
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

experiences. However, some scholars point out that the application of AI presidential election based on social media data (Ghahari et al., 2021;
technology may affect turnover intention (Li et al., 2019a). Liu et al., 2021b); to explore the political tendency towards certain
phenomena or in certain domains (e.g., Brexit, AI, data science, and
4.9. Population geography organic waste flows) (Bastos and Mercea, 2018; Folgado and Sanz, 2022;
Morone et al., 2021); to address political issues including the radical
The application of GeoAI in population geography focuses on pre­ right (Jambrina-Canseco, 2023); and to visualize or predict political
dicting population growth and population movement (e.g., Chen et al., activities (e.g., protect incidents and terrorism incidents) (Bekker, 2022;
2018; Ullah et al., 2019), disaggregating population counts based on Hao et al., 2019). Although such applications are largely limited to
other data sources to generate fine-grained population data that can be coping with texts and qualitative contents, it showcases the great po­
used for other research purposes (e.g., Khan et al., 2021; Monteiro et al., tential of GeoAI in the political implications and helps government hear
2019; Ye et al., 2019), measuring and characterising human activities the voice of the ‘silent majority’ which was usually difficult to be
(Hu et al., 2016b), delineating human settlement in the exposure to monitored via qualitative methods.
natural hazards (Herfort et al., 2019), forecasting the mortality of
populations (Perla et al., 2021) and exploring its potential driving fac­ 4.12. Regional geography
tors (Boumezoued and Elfassihi, 2021), tracking human trajectories and
mobility (Hu, 2020), and inferring people’s demographic and socio­ The application of GeoAI in regional geography is relatively limited
economic status (Zhang et al., 2020e). Such studies predominantly uti­ given the aim of regional geography is to study the interaction of
lise machine learning models (e.g., random forest, gradient boosting different cultural and natural geofactors in a specific land or regional
models) to predict the non-linear relationship between population (e.g., landscape and compare such differences across regions—it largely
counts, their demographic and socioeconomic characteristics) and other overlaps with studies in other subdomains, leading to a few papers that
phenomena (e.g., urbanisation, land use change, green space, and fall into this category as the primary subdomain. Despite the nature of
climate change). The applications of more advanced deep learning mixture in regional geographic studies, the GeoAI applications have
models are employed to derive measures from diverse data sources (e.g., been used to evaluate regional carrying capacity (Chen et al., 2011), the
classification based on remote sensing imageries or official data) that inequality and neighbourhood mixing across metropolitan areas (Kane
can be further used for the disaggregation or prediction of population and Hipp, 2019), contrasting landscapes (Hernandez-Moreno et al.,
data at different spatial scales. Common findings from these studies 2021), regional sustainability across metropolitan areas (Liu et al.,
include that GeoAI provides sufficient approaches with sound modelling 2021a), regional development (Lai et al., 2022), and regional inequality
performance to capture the complex relationship between human and of neighbourhood typologies (Lynge et al., 2022). Findings from these
the environment and to produce reliable population data that can be studies show that GeoAI applications provide highly accurate modelling
calibrated and validated by other data sources. results to explore the non-linear relationship among regional charac­
teristics and dependent phenomena and to reveal its great potential to be
4.10. Cultural geography used in comparative studies across different geographic contexts.

GeoAI has been used to advance the methodology used in the cultural 4.13. Rural geography
geography studies which are traditionally rooted in qualitative investi­
gation. For example, GoeAI is applied to processing and analysing geo­ GeoAI is increasingly becoming a vital technique in rural geography,
tagged photos and street-level images to assess ecosystem cultural offering innovative solutions to the unique challenges faced by rural
services (nonmaterial benefits people obtain from ecosystems that affect areas. One of its key applications is agricultural land use and crop
their physical and mental states(Cardoso et al., 2022; Fish et al., 2016; monitoring, where it leverages artificial intelligence technologies for
Marine et al., 2022; Richards and Tunçer, 2018) and to understand the ensuring food security (Yang et al., 2022a) and optimising urban-
spatiotemporal pattern of linguistic variations in one community (Hong, agricultural-ecological space (Wang et al., 2022a; Zeng et al., 2022).
2020). The approach of GeoAI makes the analysis more efficient by GeoAI also plays a crucial role in rural infrastructure planning and
reducing the error and time from previous manual labour work. GeoAI development by assessing and prioritizing infrastructure needs, such as
has also helped to understand the development of linguistics by recog­ demand for public transportation (Bakdur et al., 2021), the severity of
nizing handwritten words from Bangla word images (Das et al., 2020) accidents prediction (Habibzadeh et al., 2022), and policy guarantee
and post-correcting Optical Character Recognition in Hebrew (Suissa mechanism assessment (Jin et al., 2021). This contributes to more
et al., 2022). These approaches create a stable and generalized system equitable and sustainable rural development. In addition, advanced
for word recognition and correction and obtained a high accuracy rate. machine learning algorithms are employed to model and predict rural
Many studies also use social media data to understand cultural practices economic development (Khalaf et al., 2022; Qin et al., 2020; Xie et al.,
and beliefs, including measuring food cultural differences (Zhang et al., 2022) and rural population dynamics (Grossman et al., 2022; Lee,
2020c), detecting changes in perception towards smart cities (Yue et al., 2022). Furthermore, GeoAI has shown remarkable success in identifying
2022), recognizing human daily activity (Gong et al., 2019), and land use and land cover changes in rural areas that might be overlooked
measuring social-spatial boundaries (Rahimi et al., 2018). GoeAI is used by traditional remote sensing methods (Killeen et al., 2022; Saha et al.,
for data extraction, text pre-processing, and sentimental analysis in 2022; Xu et al., 2019). These insights guide rural development policies
those studies. and strategies, helping address key challenges in rural areas. Overall, the
integration of GeoAI in rural geography has the potential to revolu­
4.11. Political geography tionize the field, offering new perspectives and data-driven approaches
for the sustainable development of rural communities.
Compared to other subdomains, the application of GeoAI in political
geography is relatively limited, possibly due to the nature of political 5. Discussion: Future directions and challenges
geographic studies being more qualitative. In very recent years, there
are also some GeoAI applications, predominantly natural language 5.1. Cross-disciplinary research opportunities and beyond
processing models, that have been used to process the contents of pol­
icies, newspapers, and planning documents (Brinkley and Stahmer, Human geography has been revolutionised to be increasingly sup­
2021); to monitor people’s perception towards political initiatives (e.g., ported by spatiotemporal big data, more robust in research design to
smart city concepts) (Yigitcanlar et al., 2021); to forecast corruption and address the non-linear complex relationship between human society and

11
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

its potential drivers, more diverse in empirical studies and in turn to 5.3. Sustainable GeoAI: repeatable, reproducible and expandable
advance theoretical foundations. We have observed that the field of
GeoAI presents numerous cross-disciplinary research opportunities to GeoAI is crucial for environmental and sustainability issues, pro­
link human geography to public health, environmental science, medical cessing geospatial data efficiently. Its solutions must be repeatable,
science, decision and policy-making, and industrial practices more reproducible, and expandable for consistent, transferable, and scalable
broadly. In particular, human geography subdomains that were deeply methodologies. Repeatability requires GeoAI to provide consistent re­
rooted in social science in the past, including cultural, historical and sults with the same data and methods, calling for well-documented al­
political geography, have been empowered by GeoAI and spatiotem­ gorithms and transparent workflows. Reproducibility, achieved through
poral big data and crowdsourcing data (e.g., social media data) to varied data sets or environments, benefits from open-source platforms,
broaden its research impact and the coverage of empirical con­ standardized data, and shared code, enhancing collaboration and
texts—that can be rarely achieved by using small data (e.g., question­ progress. Expandability enables adaptation to larger or new data sets
naires) and qualitative methods. Besides, there is still so much potential and evolving queries, utilizing modular designs, cloud computing, and
to be realized by leveraging the power of GeoAI in conjunction with advanced algorithms for big data, ensuring GeoAI solutions remain
broader fields in social science such as psychology, sociology, and an­ flexible and applicable. In addition, the complexity of GeoAI methods
thropology to analyse and predict human behaviour—additionally, to and their execution environments affects their repeatability, reproduc­
enhance the studies on human-environment interactions given human ibility, and expandability. Three main approaches for developing and
behaviour was thought to be mediated through the surrounding envi­ executing GeoAI models include 1) using existing GIS or analysis soft­
ronment where they reside (e.g., urban built environment and natural ware like Geoda, ArcGIS, and others, which is user-friendly but less
environment) (Wang et al., 2023; Wang et al., 2021b). On the practical reproducible and scalable; 2) developing and executing complex models
end, GeoAI helps to achieve decision-making and evaluate different through code, which enhances reproducibility and scalability, especially
scenarios of policies in the initiatives of smart and healthy cities as well with Jupyter Notebooks. However, Trisovic et al. (2022) conduct a
as citizen participation in urban planning and design—where public large-scale study on research code quality and execution, and found that
sectors and authorities could rely on the quantitative results simulated 74 % of data science research code failed execution tests (3) utilizing
by GeoAI to optimise policy implementation and reduce social and visual programming tools like ArcGIS’s Model Builder, QGIS, Knime,
financial costs. The possibilities for cross-disciplinary research are Orange3, and Alteryx to develop and execute GeoAI models in an
endless, and the potential benefits are significant with the invention of executable workflow, reducing the burden on researchers and
new technology such as generative AI, digital twins, knowledge graphs, improving understandability, it provides a promising idea to improve
5G, and the Internet of Things (Zhang et al., 2022) to help both re­ repeatability, reproducibility, and expandability of GeoAI modes.
searchers and policymakers gain a deeper understanding of complex However, there are still relatively few applications for this at present.
urban systems and make more informed decisions that positively link Sustainable GeoAI requires a holistic approach to ensure solutions are
the academic outcomes to the real world. repeatable, reproducible, and expandable. By focusing on these aspects,
the GeoAI community can foster long-term, impactful, and collaborative
5.2. Emerging spatiotemporal data, and its issues and challenges work, contributing to a more sustainable future.

Our review finds out that human geography studies have been much 5.4. Human-centred GeoAI in the era of artificial generative intelligence
advanced by emerging spatiotemporal big data that enables geographers
to track, monitor, and quantify complex human behaviors in a large AI has witnessed significant advancement with the emergence of
spatial and temporal scale. It further indicates that the AI’s extensive Artificial General Intelligence (AGI) representing AI system’s capable of
role in human geography, yet highlights the limited interpretability of performing intellectual tasks better that a human, and sometimes
current models, a concern raised in recent studies (Hsu and Li, 2023; Liu exceeding human intelligence (e.g., ChatGPT). Leveraging AGI in
et al., 2023; Xing and Sieber, 2023). This issue, along with the potential human-centered GeoAI has the potential to deliver substantial advan­
bias in GeoAI models, is crucial for future research. For instance, models tages, including bolstering decision-making processes, optimizing
predicting human perceptions from imagery, often developed through resource management, and enhancing disaster response and recovery
broad surveys, may not suit specific geographic needs (Kang et al., efforts. The active involvement of stakeholders in the design and
2023). Additionally, these models typically require substantial, region- development of AI solutions is anticipated to enhance trust and accep­
specific data, presenting challenges in data availability and applica­ tance among the community, guaranteeing that the developed solutions
tion across different areas. Addressing these data limitations and biases are customized to cater to their specific interests and requirements. This
in GeoAI is vital for advancing the field. Our review (Fig. 6) has further approach fosters inclusivity, allowing for a sense of ownership among
found that the used data spans a variety of types and comes from a va­ the stakeholders, thus creating a collaborative atmosphere conducive to
riety of sources, e.g. authoritative data from governments to crowd­ developing efficient AI solutions. The integration of AGI in geospatial
sourced instances. This matter requires attention, as it has implications analysis raises concerns about privacy violations and deepening in­
for the quality and downstream analyses. Much of the data used in the equalities. Geospatial data, potentially misused by public or private
reviewed papers are obtained from OpenStreetMap, the freely editable entities, may infringe on privacy and liberty. AGI, relying on historical
map of the world, which in some geographies offers suitable data for data, might perpetuate biases and discrimination, aggravating societal
some of the use cases covered in this review. However, data remains inequalities and marginalizing communities. We argue that human-
highly heterogeneous (Biljecki et al., 2023) and in some locations, it may centered GeoAI must focus on transparency and accountability, allow­
not be sufficient or even detrimental for analyses. We call attention to ing stakeholders complete access to AGI-related information. It’s vital
pay attention to the quality of the data, especially those derived from a that AGI be inclusive and equitable, avoiding the reinforcement of
crowdsourced provenance. Further, a potential research direction is existing biases in geospatial analysis. Achieving this involves fair data
consideration of the impact of the propagation of errors on the outcome practices and actively reducing data biases, thus building trust and
of an analysis. Spatial data quality, a topic often ignored in human ge­ contributing to a more equitable, sustainable future through trans­
ography, regards multiple elements such as completeness, positional parent, accountable, and inclusive AGI solutions.
accuracy, and thematic accuracy (Hou and Biljecki, 2022). The quality
of each of these elements impacts different use cases in different ways, 5.5. Computational capacity subject to the quantum revolution
and it would require extensive research on understanding the reliability
of an analysis based on the input dataset. GeoAI has experienced rapid growth in recent years, yet it still faces

12
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

several geo-computation challenges that must be tackled to ensure its Conceptualization. Mengxi Zhang: Writing – review & editing, Writing
long-term success. One significant challenge is scalable, efficient, and – original draft, Resources, Investigation, Data curation. Filip Biljecki:
cost-effective data storage solutions for the large volumes of high- Writing – review & editing, Writing – original draft, Visualization,
resolution and real-time geospatial data generated by remote sensing Validation, Software, Methodology, Formal analysis, Conceptualization.
platforms, IoT devices, social media sources, and more. In addition to Tao Hu: Writing – review & editing, Writing – original draft, Data
traditional distributed and cloud-based solutions, edge cloud-based curation. Xiaokang Fu: Writing – review & editing, Writing – original
storage system provides an innovative way to improve the perfor­ draft, Data curation. Lingbo Liu: Writing – review & editing, Writing –
mance, efficiency, and scalability of geospatial big data storage by original draft, Data curation. Xintao Liu: Writing – review & editing,
reducing latency, improving bandwidth utilization, and providing high Writing – original draft, Data curation. Ruomei Wang: Resources, Data
scalability. With the development of blockchain (Zheng et al., 2018), curation. Yuanyuan Huang: Resources, Data curation. Jingjing Yan:
Geochain has the potential to significantly impact geospatial data stor­ Resources, Data curation. Jinghan Jiang: Resources, Data curation.
age by offering a decentralized, secure, and transparent solution for Michaelmary Chukwu: Resources, Data curation. Seyed Reza
managing geospatial data (Kamel Boulos et al., 2018; Mao and Golab, Naghedi: Resources, Data curation. Moein Hemmati: Resources, Data
2023). Analysing large geospatial datasets can be computationally curation. Yaxiong Shao: Resources, Data curation. Nan Jia: Resources,
intensive, particularly when using advanced machine learning algo­ Data curation. Zhiyang Xiao: Resources, Data curation. Tian Tian:
rithms that require significant processing power. Developing scalable Resources, Data curation. Yaxin Hu: Resources, Data curation. Lix­
processing techniques that can handle large datasets without compro­ iaona Yu: Resources, Data curation. Winston Yap: Resources, Data
mising accuracy is crucial. This may involve parallel processing, curation. Edgardo Macatulad: Resources, Data curation. Zhuo Chen:
distributed computing, or leveraging specialized hardware, such as Resources, Data curation. Yunhe Cui: Resources, Data curation. Koichi
GPUs, to improve the efficiency of GeoAI applications. Quantum Ito: Resources, Data curation. Mengbi Ye: Resources, Data curation.
computing (Steane, 1998), an emerging technology that exploits the Zicheng Fan: Resources, Data curation. Binyu Lei: Resources, Data
principles of quantum mechanics, has the potential to revolutionize curation. Shuming Bao: Resources, Data curation.
various fields, including GeoAI. Although still in its early stages, quan­
tum computing could offer significant advantages by enhancing Declaration of competing interest
computational power, processing real-time geospatial big data,
improving machine learning algorithms (Riedel et al., 2021), and solv­ The authors declare that they have no known competing financial
ing complex optimization problems (e.g., NP-hard problems) (Werner, interests or personal relationships that could have appeared to influence
2019). the work reported in this paper.

6. Conclusion Data availability

Human geography has undergone a transformative shift, increas­ No data was used for the research described in the article.
ingly relying on spatiotemporal big data to enhance research design and
address the intricate, non-linear relationships between human society Acknowledgement
and its potential drivers. This evolution is marked by a greater diversity
in empirical studies, contributing to the advancement of theoretical This research is jointly supported by the US National Science Foun­
foundations. Our review reveals that the integration of emerging dation Awards (#1841403); the Japan Society for the Promotion of
spatiotemporal big data has significantly propelled human geography Science KAKENHI research grant (JP22F21725); the project Large-scale
studies, allowing geographers to track, monitor, and quantify complex 3D Geospatial Data for Urban Analytics, which is supported by the Na­
human behaviors on a large spatial and temporal scale. The intersection tional University of Singapore (NUS) under the Start Up Grant R-295-
of GeoAI with quantum computing is poised to revolutionize human 000-171-133; the project Multi-scale Digital Twins for the Urban Envi­
geography studies further, providing advanced tools to simulate spatial ronment: From Heartbeats to Cities, which is supported by the Singapore
phenomena and enhance predictions related to the environment and Ministry of Education Academic Research Fund Tier 1; the Singapore
population dynamics. This integration will empower researchers to International Graduate Award (SINGA) scholarship provided by the
process and analyze extensive datasets at an unprecedented speed, Agency for Science, Technology, and Research (A*STAR) and the NUS;
enabling a more detailed exploration of spatial relationships. It is the NUS President’s Graduate Fellowship (GRSUP0600047 PGF NUSGS
essential, however, to approach the development and use of GeoAI CDE-IS); the Foreign PhD Scholarship Grant of the Department of Sci­
responsibly and ethically, considering the potential social and environ­ ence and Technology - Engineering Research and Development for
mental impacts of its implementation. We advocate for collaborative Technology, Philippines; the NUS Graduate Research Scholarship
efforts across disciplines and sectors, involving government entities, (NUSGS.031/21).
public and private authorities, and academia. These concerted actions
will contribute to enriching the roadmap of GeoAI in human geography, Appendix A. Supplementary material
extending its application to broader geographic paradigms. This, in turn,
will empower our geographers to seize research opportunities and Supplementary data to this article can be found online at https://doi.
leverage insights from the emerging data and AI deluge. org/10.1016/j.jag.2024.103734.

CRediT authorship contribution statement

References

Siqin Wang: Writing – review & editing, Writing – original draft, Acheampong, A.O., Boateng, E.B., 2019. Modelling carbon emission intensity:
Visualization, Validation, Supervision, Software, Resources, Project Application of artificial neural network. J. Clean. Prod. 225, 833–856.
Agonafir, C., Lakhankar, T., Khanbilvardi, R., Krakauer, N., Radell, D., Devineni, N.,
administration, Methodology, Investigation, Funding acquisition,
2022. A machine learning approach to evaluate the spatial variability of New York
Formal analysis, Data curation, Conceptualization. Xiao Huang: City’s 311 street flooding complaints. Comput. Environ. Urban Syst. 97, 101854.
Writing – review & editing, Writing – original draft, Validation, Re­ Ahmed, I., Ahmad, M., Rodrigues, J.J., Jeon, G., Din, S., 2021. A deep learning-based
sources, Methodology, Investigation, Funding acquisition, Formal social distance monitoring framework for COVID-19. Sustain. Cities Soc. 65, 102571.
Alastal, A.I., Shaqfa, A.H., 2022. Geoai technologies and their application areas in urban
analysis, Data curation, Conceptualization. Pengyuan Liu: Writing – planning and development: concepts, opportunities and challenges in smart city
review & editing, Writing – original draft, Methodology, Data curation, (Kuwait, study case). J. Data Anal. Inform. Process. 10, 110–126.

13
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Alizadeh, M., Alizadeh, E., Asadollahpour Kotenaee, S., Shahabi, H., Beiranvand D’Uva, D., Rolando, A., 2023. A Method to Select and Optimize Slow Tourism Routes
Pour, A., Panahi, M., Bin Ahmad, B., Saro, L., 2018. Social vulnerability assessment Using a Quality Index Procedure Based on Image Segmentation and DTM Modelling
using artificial neural network (ANN) model for earthquake hazard in Tabriz city, Based on NURBS: The Case Study of Multimodal Access to Inner Places from the
Iran. Sustainability 10, 3376. Nodes of the Adriatic Coastline’s Infrastructure Bundle. Sustainability 15, 373.
AL-Musaylh, M.S., Al-Daffaie, K., Prasad, R., 2021. Gas consumption demand forecasting Das, D., Nayak, D.R., Dash, R., Majhi, B., Zhang, Y.D., 2020. H-WordNet: a holistic
with empirical wavelet transform based machine learning model: A case study. Int. J. convolutional neural network approach for handwritten word recognition. IET
Energy Res. 45, 15124–15138. Image Proc. 14, 1794–1805.
Amiruzzaman, M., Curtis, A., Zhao, Y., Jamonnak, S., Ye, X., 2021. Classifying crime Debnath, R., Bardhan, R., Darby, S., Mohaddes, K., Sunikka-Blank, M., Coelho, A.C.V.,
places by neighborhood visual appearance and police geonarratives: A machine Isa, A., 2021. Words against injustices: A deep narrative analysis of energy cultures
learning approach. J. Comput. Soc. Sci. 1–25. in poverty of Abuja, Mumbai and Rio de Janeiro. Energy Res. Soc. Sci. 72, 101892.
Andrade, R., Alves, A., Bento, C., 2020. POI mining for land use classification: A case Demetriou, D., Stillwell, J., See, L., 2012. Land consolidation in Cyprus: why is an
study. ISPRS Int. J. Geo Inf. 9, 493. integrated planning and decision support system required? Land Use Policy 29,
Apollonio, C., Balacco, G., Novelli, A., Tarantino, E., Piccinni, A.F., 2016. Land use 131–142.
change impact on flooding areas: The case study of Cervaro Basin (Italy). Dias, E.G., Oliveira, L.K.d., Isler, C.A., 2021. Assessing the effects of delivery attributes
Sustainability 8, 996. on e-shopping consumer behaviour. Sustainability 14, 13.
Auerbach, J., Blackburn, C., Barton, H., Meng, A., Zegura, E., 2020. Coupling data Ding, C., Cao, X.J., Næss, P., 2018. Applying gradient boosting decision trees to examine
science with community crowdsourcing for urban renewal policy analysis: An non-linear effects of the built environment on driving distance in Oslo. Transport.
evaluation of Atlanta’s Anti-Displacement Tax Fund. Environ.. Plann. b: Urban Anal. Res. Part A: Policy Pract. 110, 107–117.
City Sci. 47, 1081–1097. Diou, C., Lelekas, P., Delopoulos, A., 2018. Image-based surrogates of socio-economic
Bakdur, A., Masui, F., Ptaszynski, M., 2021. Predicting Increase in demand for public status in urban neighborhoods using deep multiple instance learning. J. Imaging 4,
buses in university students daily life needs: case study based on a city in Japan. 125.
Sustainability 13, 5137. Doi, S., Mizuno, T., Fujiwara, N., 2021. Estimation of socioeconomic attributes from
Bao, J., Liu, P., Ukkusuri, S.V., 2019. A spatiotemporal deep learning approach for location information. J. Comput. Soc. Sci. 4, 187–205.
citywide short-term crash risk prediction with multi-source data. Accid. Anal. Prev. Fang, Z., Jin, Y., Yang, T., 2022. Incorporating planning intelligence into deep learning:
122, 239–254. A planning support tool for street network design. J. Urban Technol. 29, 99–114.
Bastos, M., Mercea, D., 2018. Parametrizing Brexit: mapping Twitter political space to Feldmeyer, D., Meisch, C., Sauter, H., Birkmann, J., 2020. Using OpenStreetMap data
parliamentary constituencies. Inf. Commun. Soc. 21, 921–939. and machine learning to generate socio-economic indicators. ISPRS Int. J. Geo Inf. 9,
Bekker, M., 2022. Depends on how you count them: the value of general propensity 498.
choropleth maps for visualising databases of protest incidents. J. Maps 1–8. Feng, Y., Sester, M., 2018. Extraction of pluvial flood relevant volunteered geographic
Ben Zion, E., Lerner, B., 2018. Identifying and predicting social lifestyles in people’s information (VGI) by deep learning from user generated texts and photos. ISPRS Int.
trajectories by neural networks. EPJ Data Sci. 7, 1–27. J. Geo Inf. 7, 39.
Biljecki, F., Chow, Y.S., Lee, K., 2023. Quality of crowdsourced geospatial building Fish, R., Church, A., Winter, M., 2016. Conceptualising cultural ecosystem services: A
information: A global assessment of OpenStreetMap attributes. Build. Environ. novel framework for research and critical engagement. Ecosyst. Serv. 21, 208–217.
Boumezoued, A., Elfassihi, A., 2021. Mortality data correction in the absence of monthly Folgado, M.G., Sanz, V., 2022. Exploring the political pulse of a country using data
fertility records. Ins: Mathematics Econ 99, 486–508. science tools. J. Comput. Soc. Sci. 5, 987–1000.
Brinkley, C., Stahmer, C., 2021. What is in a plan? Using natural language processing to Fontes, T., Arantes, M., Figueiredo, P.V., Novais, P., 2022. A cluster-based approach
read 461 California city general plans. J. Plan. Educ. Res. using smartphone data for bike-sharing docking stations identification: Lisbon case
Brown, R.A., Dickerson, D.L., Klein, D.J., Agniel, D., Johnson, C.L., D’Amico, E.J., 2021. study. Smart Cities 5, 251–275.
Identifying as American Indian/Alaska Native in urban areas: Implications for Furgała-Selezniow, G., Jankun-Woźnicka, M., Kruk, M., Omelan, A.A., 2021. Land use
adolescent behavioral health and well-being. Youth Soc. 53, 54–75. and land cover pattern as a measure of tourism impact on a lakeshore zone. Land 10,
Bui, D.T., Hoang, N.-D., Samui, P., 2019. Spatial pattern analysis and prediction of forest 787.
fire using new machine learning approach of Multivariate Adaptive Regression Ghahari, S., Queiroz, C., Labi, S., McNeil, S., 2021. Cluster forecasting of corruption
Splines and Differential Flower Pollination optimization: A case study at Lao Cai using nonlinear autoregressive models with exogenous variables (NARX)—An
province (Viet Nam). J. Environ. Manage. 237, 476–487. artificial neural network analysis. Sustainability 13, 11366.
Cardoso, A.S., Renna, F., Moreno-Llorca, R., Alcaraz-Segura, D., Tabik, S., Ladle, R.J., Ghahramani, M., Galle, N.J., Ratti, C., Pilla, F., 2021b. Tales of a city: Sentiment analysis
Vaz, A.S., 2022. Classifying the content of social media images to support cultural of urban green space in Dublin. Cities 119, 103395.
ecosystem service assessments using deep learning models. Ecosyst. Serv. 54, Ghahramani, M., Pilla, F., 2021. Leveraging artificial intelligence to analyze the COVID-
101410. 19 distribution pattern based on socio-economic determinants. Sustain. Cities Soc.
Čeh, M., Kilibarda, M., Lisec, A., Bajat, B., 2018. Estimating the performance of random 69, 102848.
forest versus multiple regression for predicting prices of the apartments. ISPRS Int. J. Ghahramani, M., Galle, N.J., Duarte, F., Ratti, C., Pilla, F., 2021a. Leveraging artificial
Geo Inf. 7, 168. intelligence to analyze citizens’ opinions on urban green space. City Environ.
Cesare, N., Dwivedi, P., Nguyen, Q.C., Nsoesie, E.O., 2019. Use of social media, search Interactions 10, 100058.
queries, and demographic data to assess obesity prevalence in the United States. Giglio, S., Bertacchini, F., Bilotta, E., Pantano, P., 2019. Using social media to identify
Palgrave Communications 5, 1–9. tourism attractiveness in six Italian cities. Tour. Manag. 72, 306–312.
Chang, H., Li, L., Huang, J., Zhang, Q., Chin, K.-S., 2022. Tracking traffic congestion and Golder, S., Klein, A.Z., Magge, A., O’Connor, K., Cai, H., Weissenbacher, D., Gonzalez-
accidents using social media data: A case study of Shanghai. Accid. Anal. Prev. 169, Hernandez, G., 2020. Extending A chronological and geographical analysis of
106618. personal reports of COVID-19 on Twitter to England, UK. medRxiv.
Chang, X., Wu, J., Liu, H., Yan, X., Sun, H., Qu, Y., 2019. Travel mode choice: a data Gong, J., Li, R., Yao, H., Kang, X., Li, S., 2019. Recognizing human daily activity using
fusion model using machine learning methods and evidence from travel diary survey social media sensors and deep learning. Int. J. Environ. Res. Public Health 16, 3955.
data. Transport. A: Transp. Sci. 15, 1587–1612. Gregory, D., Johnston, R., Pratt, G., Watts, M., Whatmore, S., 2011. The dictionary of
Chen, P.C., Hsieh, H.Y., Su, K.W., Sigalingging, X.K., Chen, Y.R., Leu, J.S., 2020a. human geography. John Wiley & Sons.
Predicting station level demand in a bike-sharing system using recurrent neural Grossman, I., Bandara, K., Wilson, T., Kirley, M., 2022. Can machine learning improve
networks. IET Intel. Transport Syst. 14, 554–561. small area population forecasts? A forecast combination approach. Comput. Environ.
Chen, Y., Liu, X., Li, X., Liu, Y., Xu, X., 2016. Mapping the fine-scale spatial pattern of Urban Syst. 95, 101806.
housing rent in the metropolitan area by using online rental listings and ensemble Guo, Q., He, Z., 2021. Prediction of the confirmed cases and deaths of global COVID-19
learning. Appl. Geogr. 75, 200–212. using artificial intelligence. Environ. Sci. Pollut. Res. 28, 11672–11682.
Chen, J., Pei, T., Shaw, S.-L., Lu, F., Li, M., Cheng, S., Liu, X., Zhang, H., 2018. Fine- Guo, P., Liu, T., Zhang, Q., Wang, L., Xiao, J., Zhang, Q., Luo, G., Li, Z., He, J., Zhang, Y.,
grained prediction of urban population using mobile phone location data. Int. J. 2017. Developing a dengue forecast model using machine learning: A case study in
Geogr. Inf. Sci. 32, 1770–1786. China. PLoS Negl. Trop. Dis. 11, e0005973.
Chen, T.-H.-K., Qiu, C., Schmitt, M., Zhu, X.X., Sabel, C.E., Prishchepov, A.V., 2020b. Habibzadeh, M., Ameri, M., Ziari, H., Kamboozia, N., Sadat Haghighi, S.M., 2022.
Mapping horizontal and vertical urban densification in Denmark with Landsat time- Presentation of Machine Learning Approaches for Predicting the Severity of
series from 1985 to 2018: A semantic segmentation solution. Remote Sens. Environ. Accidents to Propose the Safety Solutions on Rural Roads. J Adv Transport.
251, 112096. Handayanto, R.T., Tripathi, N.K., Kim, S.M., Guha, S., 2017. Achieving a sustainable
Chen, J.-P., Zeng, M., Duan, Y.-J., 2011. Regional carrying capacity evaluation and urban form through land use optimisation: insights from Bekasi City’s land-use plan
prediction based on GIS in the Yangtze River Delta, China. Int. J. Geogr. Inf. Sci. 25, (2010–2030). Sustainability 9, 221.
171–190. Hao, M., Jiang, D., Ding, F., Fu, J., Chen, S., 2019. Simulating spatio-temporal patterns of
Choi, Y.S., Kim, H., Sohn, D., 2022. Mapping Social Distress: A Computational Approach terrorism incidents on the Indochina Peninsula with GIS and the random forest
to Spatiotemporal Distribution of Anxiety. Soc. Sci. Comput. Rev. 40, 598–617. method. ISPRS Int. J. Geo Inf. 8, 133.
Choudhury, T., Arunachalam, R., Khanna, A., Jasinska, E., Bolshev, V., Panchenko, V., Helbich, M., Poppe, R., Oberski, D., van Emmichoven, M.Z., Schram, R., 2021. Can’t see
Leonowicz, Z., 2022. A Social Network Analysis Approach to COVID-19 Community the wood for the trees? An assessment of street view-and satellite-derived greenness
Detection Techniques. Int. J. Environ. Res. Public Health 19, 3791. measures in relation to mental health. Landsc. Urban Plan. 214, 104181.
Cole, M.A., Elliott, R.J., Liu, B., 2020. The impact of the Wuhan Covid-19 lockdown on Heredia, C., Moreno, S., Yushimito, W.F., 2021. Characterization of mobility patterns
air pollution and health: a machine learning and augmented synthetic control with a hierarchical clustering of origin-destination gps taxi data. IEEE Trans. Intell.
approach. Environ. Resour. Econ. 76, 553–580. Transp. Syst. 23, 12700–12710.
Crivellari, A., Beinat, E., 2020. LSTM-based deep learning model for predicting
individual mobility traces of short-term foreign tourists. Sustainability 12, 349.

14
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Herfort, B., Li, H., Fendrich, S., Lautenbach, S., Zipf, A., 2019. Mapping human Lavelle-Hill, R., Smith, G., Mazumder, A., Landman, T., Goulding, J., 2021. Machine
settlements with higher accuracy and less volunteer efforts by combining learning methods for “wicked” problems: exploring the complex drivers of modern
crowdsourcing and deep learning. Remote Sens. (Basel) 11, 1799. slavery. Human. Soc. Sci. Commun. 8, 1–11.
Hernandez-Moreno, A., Echeverria, C., Sotomayor, B., Soto, D.P., 2021. Relationship Lee, H., 2022. What drives the performance of Chinese urban and rural secondary
between anthropization and spatial patterns in two contrasting landscapes of Chile. schools: A machine learning approach using PISA 2018. Cities 123, 103609.
Appl. Geogr. 137, 102599. Lee, A.-C.-D., Rinner, C., 2015. Visualizing urban social change with self-organizing
Hipp, J.R., Kane, K., Kim, J.H., 2017. Recipes for neighborhood development: A machine maps: Toronto neighbourhoods, 1996–2006. Habitat Int. 45, 92–98.
learning approach toward understanding the impact of mixing in neighborhoods. Leszczynski, A., Crampton, J., 2016. Introduction: Spatial big data and everyday life. Big
Landsc. Urban Plan. 164, 1–12. Data Society 3.
Hoggart, K., 2002. Researching human geography. Levers, C., Schneider, M., Prishchepov, A.V., Estel, S., Kuemmerle, T., 2018. Spatial
Hong, S.-Y., 2020. Linguistic landscapes on street-level images. ISPRS Int. J. Geo Inf. 9, variation in determinants of agricultural land abandonment in Europe. Sci. Total
57. Environ. 644, 95–111.
Hou, Y., Biljecki, F., 2022. A comprehensive framework for evaluating the quality of Li, J.J., Bonn, M.A., Ye, B.H., 2019a. Hotel employee’s artificial intelligence and robotics
street view imagery. Int. J. Appl. Earth Obs. Geoinf. 115, 103094. awareness and its impact on turnover intention: The moderating roles of perceived
Hou, Y., Zhang, K., Li, G., 2021. Service robots or human staff: How social crowding organizational support and competitive psychological climate. Tour. Manag. 73,
shapes tourist preferences. Tour. Manag. 83, 104242. 172–181.
Hu, B., 2020. Trajectories of informal care intensity among the oldest-old Chinese. Soc Li, W., Hsu, C.-Y., 2022. GeoAI for large-scale image analysis and machine vision: Recent
Sci Med 266, 113338. progress of artificial intelligence in geography. ISPRS Int. J. Geo Inf. 11, 385.
Hu, N., Legara, E.F., Lee, K.K., Hung, G.G., Monterola, C., 2016a. Impacts of land use and Li, Y., Liu, C., Gao, Q., Wu, D., Li, F., Du, Y., 2022a. ConTrack distress dataset: a
amenities on public transport use, urban planning and design. Land Use Policy 57, continuous observation for pavement deterioration spatio-temporal analysis. IEEE
356–367. Trans. Intell. Transp. Syst. 23, 25004–25017.
Hu, Z., Wang, Y., Liu, Y., Long, H., Peng, J., 2016b. Spatio-temporal patterns of urban- Li, M., Sheng, H., Irvin, J., Chung, H., Ying, A., Sun, T., Ng, A.Y., Rodriguez, D.A., 2023a.
rural development and transformation in east of the “Hu Huanyong Line”, China. Marked crosswalks in US transit-oriented station areas, 2007–2020: A computer
ISPRS Int. J. Geo-Inform. 5, 24. vision approach using street view imagery. Environ.. Plann. b: Urban Anal. City Sci.
Hu, Z., Zhou, J., Huang, K., Zhang, E., 2022. A data-driven approach for traffic crash 50, 350–369.
prediction: a case study in Ningbo, China. Int. J. Intell. Transp. Syst. Res. 20, Li, Z., Wei, H., Wu, Y., Su, S., Wang, W., Qu, C., 2019b. Impact of community deprivation
508–518. on urban park access over time: Understanding the relative role of contributors for
Huang, Z., Wang, D., Yin, Y., Li, X., 2021. A spatiotemporal bidirectional attention-based urban planning. Habitat Int. 92, 102031.
ride-hailing demand prediction model: a case study in beijing during COVID-19. Li, Z., Liang, Z., Feng, L., Fan, Z., 2022b. Beyond accessibility: a multidimensional
IEEE Trans. Intell. Transp. Syst. 23, 25115–25126. evaluation of urban park equity in Yangzhou, China. ISPRS Int. J. Geo-Inform. 11,
Ibrahim, H., Khattab, Z., Khattab, T., Abraham, R., 2021. Expatriates’ housing dispersal 429.
outlook in a rapidly developing metropolis based on urban growth predicted using a Li, Y., Yabuki, N., Fukuda, T., 2023b. Integrating GIS, deep learning, and environmental
machine learning algorithm. Housing Pol Debate 1–21. sensors for multicriteria evaluation of urban street walkability. Landsc. Urban Plan.
Ivan, K., Holobâcă, I.-H., Benedek, J., Török, I., 2020. VIIRS nighttime light data for 230, 104603.
income estimation at local level. Remote Sens. (Basel) 12, 2950. Li, W., 2022. GeoAI in social science. Handbook of Spatial Analysis in the Social Sciences,
Jambrina-Canseco, B., 2023. The stories we tell ourselves: Local newspaper reporting 291-304.
and support for the radical right. Polit. Geogr. 100, 102778. Lin, S., Döngül, E.S., Uygun, S.V., Öztürk, M.B., Huy, D.T.N., Tuan, P.V., 2022. Exploring
Jiang, S., Alves, A., Rodrigues, F., Ferreira Jr, J., Pereira, F.C., 2015. Mining point-of- the relationship between abusive management, self-efficacy and organizational
interest data from social networks for urban land use classification and performance in the context of human–machine interaction technology and artificial
disaggregation. Comput. Environ. Urban Syst. 53, 36–46. intelligence with the effect of ergonomics. Sustainability 14, 1949.
Jin, X., Zuo, X., Dong, X., Dong, Y., Ding, H., 2021. Analysis of the policy guarantee Liu, P., Biljecki, F., 2022. A review of spatially-explicit GeoAI applications in Urban
mechanism of rural infrastructure based on deep learning. Technol Forecast Soc Geography. Int. J. Appl. Earth Obs. Geoinf. 112, 102936.
Change 166, 120605. Liu, H., Chen, N., Wang, X., 2021a. Comparing regional sustainability and transportation
Kadar, C., Pletikosa, I., 2018. Mining large-scale human mobility data for long-term sustainability at the metropolitan level in the US using artificial neural network
crime prediction. EPJ Data Sci. 7, 1–27. clustering techniques. Transp. Res. Rec. 2675, 1655–1669.
Kamel Boulos, M.N., Wilson, J.T., Clauson, K.A., 2018. Geospatial blockchain: promises, Liu, K., Chen, J., Li, R., Peng, T., Ji, K., Gao, Y., 2022. Nonlinear effects of community
challenges, and scenarios in health and healthcare. Int. J. Health Geogr. 17, 1–10. built environment on car usage behavior: a machine learning approach.
Kane, K., Hipp, J.R., 2019. Rising inequality and neighbourhood mixing in US metro Sustainability 14, 6722.
areas. Reg. Stud. 53, 1680–1695. Liu, R., Yao, X., Guo, C., Wei, X., 2021b. Can we forecast presidential election using
Kang, Y., Cho, N., Yoon, J., Park, S., Kim, J., 2021. Transfer learning of a deep learning twitter data? an integrative modelling approach. Ann. GIS 27, 43–56.
model for exploring tourists’ urban image using geotagged photos. ISPRS Int. J. Geo Logan, T., Zaitchik, B., Guikema, S., Nisbet, A., 2020. Night and day: The influence and
Inf. 10, 137. relative importance of urban characteristics on remotely sensed land surface
Kang, J., Guo, X., Fang, L., Wang, X., Fan, Z., 2022. Integration of Internet search data to temperature. Remote Sens. Environ. 247, 111861.
predict tourism trends using spatial-temporal XGBoost composite model. Int. J. Luo, E., Kuffer, M., Wang, J., 2022a. Urban poverty maps-From characterising
Geogr. Inf. Sci. 36, 236–252. deprivation using geo-spatial data to capturing deprivation from space. Sustain.
Kawlra, G., Sakamoto, K., 2021. Spatialising urban health vulnerability: An analysisof Cities Soc. 84, 104033.
NYC’s critical infrastructure during COVID-19. Urban Stud.. Luo, Y., Yan, J., McClure, S.C., Li, F., 2022b. Socioeconomic and environmental factors of
Khalaf, C., Michaud, G., Jolley, G.J., 2022. Toward a new rural typology: mapping poverty in China using geographically weighted random forest regression model.
resources, opportunities, and challenges. Econ. Devel. Quart. 36, 276–293. Environ. Sci. Pollut. Res. 1–13.
Khan, N., Ullah, A., Haq, I.U., Menon, V.G., Baik, S.W., 2021. SD-Net: Understanding Lv, Y., Zhi, D., Sun, H., Qi, G., 2021. Mobility pattern recognition based prediction for the
overcrowded scenes in real-time via an efficient dilated convolutional neural subway station related bike-sharing trips. Transport. Res. Part C: Emerg. Technol.
network. J. Real-Time Image Proc. 18, 1729–1743. 133, 103404.
Ki, D., Lee, S., 2021. Analyzing the effects of Green View Index of neighborhood streets Lynge, H., Visagie, J., Scheba, A., Turok, I., Everatt, D., Abrahams, C., 2022. Developing
on walking time using Google Street View and deep learning. Landsc. Urban Plan. neighbourhood typologies and understanding urban inequality: a data-driven
205, 103920. approach. Reg. Stud. Reg. Sci. 9, 618–640.
Killeen, J., Jaupi, L., Barrett, B., 2022. Impact assessment of humanitarian demining Ma, J., Cheng, J.C., Jiang, F., Chen, W., Zhang, J., 2020. Analyzing driving factors of land
using object-based peri-urban land cover classification and morphological building values in urban scale based on big data and non-linear machine learning techniques.
detection from VHR Worldview imagery. Remote Sens. Appl.: Soc. Environ. 27, Land Use Policy 94, 104537.
100766. Mao, C., Golab, W., 2023. GeoChain: A Locality-Based Sharding Protocol for
Kim, K., 2019. Identifying the structure of cities by clustering using a new similarity Permissioned Blockchains, 24th International Conference on Distributed Computing
measure based on smart card data. IEEE Trans. Intell. Transp. Syst. 21, 2002–2011. and Networking, pp. 70-79.
Kim, J., Nirjhar, E.H., Kim, J., Chaspari, T., Ham, Y., Winslow, J.F., Lee, C., Ahn, C.R., Marine, N., Arnaiz-Schmitz, C., Santos-Cid, L., Schmitz, M.F., 2022. Can we foresee
2022. Capturing environmental distress of pedestrians using multimodal data: the landscape interest? Maximum entropy applied to social media photographs: a case
interplay of biosignals and image-based data. J. Comput. Civ. Eng. 36. study in madrid. Land 11, 715.
Knaap, E., 2017. The cartography of opportunity: Spatial data science for equitable urban Miller, H.J., 2004. Tobler’s first law and spatial analysis. Ann. Assoc. Am. Geogr. 94,
policy. Housing Pol Debate 27, 913–940. 284–289.
Kruse, J., Kang, Y., Liu, Y.-N., Zhang, F., Gao, S., 2021. Places for play: Understanding Moher, D., Liberati, A., Tetzlaff, J., Altman, D.J., Group, P., 2010. Preferred reporting
human perception of playability in cities using street view images and deep learning. items for systematic reviews and meta-analyses: the PRISMA statement. Int. J. Surg.
Comput. Environ. Urban Syst. 90, 101693. 8, 336–341.
Lai, Y., Sun, W., Schmöcker, J.-D., Fukuda, K., Axhausen, K.W., 2022. Explaining a Molina-García, J., García-Massó, X., Estevan, I., Queralt, A., 2019. Built environment,
century of Swiss regional development by deep learning and SHAP values. Environ. psychosocial factors and active commuting to school in adolescents: clustering a self-
Plann. B: Urban Analytics and City Science, 23998083221116895. organizing map analysis. Int. J. Environ. Res. Public Health 16, 83.
Larkin, A., Hystad, P., 2019. Evaluating street view exposure measures of visible green Monteiro, J., Martins, B., Murrieta-Flores, P., Pires, J.M., 2019. Spatial disaggregation of
space for health research. J. Expo. Sci. Environ. Epidemiol. 29, 447–456. historical census data leveraging multiple sources of ancillary information. ISPRS
Int. J. Geo Inf. 8, 327.

15
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Morone, P., Yilan, G., Imbert, E., 2021. Using fuzzy cognitive maps to identify better using novel ensembles of bagging algorithms. Stoch Environ. Res Risk Assess 34,
policy strategies to valorize organic waste flows: An Italian case study. J. Clean. 2277–2300.
Prod. 319, 128722. Talukdar, S., Pal, S., Singha, P., 2021. Proposing artificial intelligence based livelihood
Ning, H., Li, Z., Hodgson, M.E., Wang, C., 2020. Prototyping a social media flooding vulnerability index in river islands. J. Clean. Prod. 284, 124707.
photo screening system based on deep learning. ISPRS Int. J. Geo Inf. 9, 104. Tomar, A., Gupta, N., 2020. Prediction for the spread of COVID-19 in India and
Niu, K., Wang, C., Zhou, X., Zhou, T., 2019. Predicting ride-hailing service demand via effectiveness of preventive measures. Sci. Total Environ. 728, 138762.
RPA-LSTM. IEEE Trans. Veh. Technol. 68, 4213–4222. Tomeny, T.S., Vargo, C.J., El-Toukhy, S., 2017. Geographic and demographic correlates
Nuzzolo, A., Comi, A., 2016. Individual utility-based path suggestions in transit trip of autism-related anti-vaccine beliefs on Twitter, 2009–15. Soc. Sci. Med. 191,
planners. IET Intel. Transport Syst. 10, 219–226. 168–175.
Obringer, R., White, D.D., 2023. Leveraging unsupervised learning to develop a typology Torrens, P., Li, X., Griffin, W.A., 2011. Building agent-based walking models by machine-
of residential water users’ attitudes towards conservation. Water Resour. Manag. 37, learning on diverse databases of space-time trajectory samples. Trans. GIS 15, 67–94.
37–53. Tran, M., Draeger, C., Wang, X., Nikbakht, A., 2023. Monitoring the well-being of
Perla, F., Richman, R., Scognamiglio, S., Wüthrich, M.V., 2021. Time-series forecasting of vulnerable transit riders using machine learning based sentiment analysis and social
mortality rates using deep learning. Scand. Actuar. J. 2021, 572–598. media: Lessons from COVID-19. Environ. Plann. b: Urban Anal. City Sci. 50, 60–75.
Pierdicca, R., Paolanti, M., 2022. GeoAI: a review of artificial intelligence approaches for Tran, P.T., Zhao, M., Yamamoto, K., Minet, L., Nguyen, T., Balasubramanian, R., 2020.
the interpretation of complex geomatics data. Geosci. Instrument., Methods Data Cyclists’ personal exposure to traffic-related air pollution and its influence on
Syst. 11, 195–218. bikeability. Transport. Res. Part d: Transp. Environ. 88, 102563.
Qin, X., Li, Y., Lu, Z., Pan, W., 2020. What makes better village economic development in Trisovic, A., Lau, M.K., Pasquier, T., Crosas, M., 2022. A large-scale study on research
traditional agricultural areas of China? Evidence from 338 villages. Habitat Int. 106, code quality and execution. Sci. Data 9, 60.
102286. Ullah, S., Tahir, A.A., Akbar, T.A., Hassan, Q.K., Dewan, A., Khan, A.J., Khan, M., 2019.
Qiu, J., Jing, Y., Peng, W., Du, L., Hu, Y., 2022. Identifying critical transfer zones to Remote sensing-based quantification of the relationships between land use land
coordinate transit with on-demand services using crowdsourced trajectory data. cover changes and surface temperature over the Lower Himalayan Region.
J. Intell. Transp. Syst. 1–23. Sustainability 11, 5492.
Rafiei, M.H., Adeli, H., 2016. A novel machine learning model for estimation of sale Van Roy, V., Vertesy, D., Damioli, G., 2020. AI and robotics innovation. Handbook of
prices of real estate units. J. Constr. Eng. Manage. 142. labor, human resources population economics, 1-35.
Rahimi, S., Mottahedi, S., Liu, X., 2018. The geography of taste: using yelp to study urban van Weerdenburg, D., Scheider, S., Adams, B., Spierings, B., van der Zee, E., 2019. Where
culture. ISPRS Int. J. Geo Inf. 7, 376. to go and what to do: Extracting leisure activity potentials from Web data on urban
Ramírez, T., Hurtubia, R., Lobel, H., Rossetti, T., 2021. Measuring heterogeneous space. Comput. Environ. Urban Syst. 73, 143–156.
perception of urban space with massive data and machine learning: An application Wang, J., Biljecki, F., 2022. Unsupervised machine learning in urban studies: A
to safety. Landsc. Urban Plan. 208, 104002. systematic review of applications. Cities 129, 103925.
Rao, N.D., Ummel, K., 2017. White goods for white people? Drivers of electric appliance Wang, S., Cai, W., Tao, Y., Sun, Q.C., Wong, P.P.Y., Thongking, W., Huang, X., 2023.
growth in emerging economies. Energy Res. Soc. Sci. 27, 106–116. Nexus of heat-vulnerable chronic diseases and heatwave mediated through tri-
Reades, J., De Souza, J., Hubbard, P., 2019. Understanding urban gentrification through environmental interactions: A nationwide fine-grained study in Australia. J. Environ.
machine learning. Urban Stud. 56, 922–942. Manage. 325, 116663.
Reid, C.E., Considine, E.M., Watson, G.L., Telesca, D., Pfister, G.G., Jerrett, M., 2019. Wang, D., Fu, J., Xie, X., Ding, F., Jiang, D., 2022a. Spatiotemporal evolution of urban-
Associations between respiratory health and ozone and fine particulate matter agricultural-ecological space in China and its driving mechanism. J. Clean. Prod.
during a wildfire event. Environ. Int. 129, 291–298. 371, 133684.
Reier Forradellas, R.F., Náñez Alonso, S.L., Jorge-Vazquez, J., Rodriguez, M.L., 2020. Wang, R., Helbich, M., Yao, Y., Zhang, J., Liu, P., Yuan, Y., Liu, Y., 2019a. Urban
Applied machine learning in social sciences: neural networks and crime prediction. greenery and mental wellbeing in adults: Cross-sectional mediation analyses on
Soc. Sci. 10, 4. multiple pathways across different greenery measures. Environ. Res. 176, 108535.
Richards, D.R., Tunçer, B., 2018. Using image recognition to automate assessment of Wang, R., Liu, Y., Lu, Y., Zhang, J., Liu, P., Yao, Y., Grekousis, G., 2019b. Perceptions of
cultural ecosystem services from social media photographs. Ecosyst. Serv. 31, built environment and health outcomes for older Chinese in Beijing: A big data
318–325. approach with street view images and deep learning technique. Comput. Environ.
Riedel, M., Cavallaro, G., Benediktsson, J.A., 2021. Practice and experience in using Urban Syst. 78, 101386.
parallel and scalable machine learning in remote sensing from HPC over cloud to Wang, S., Liu, Y., Lam, J., Kwan, M.-P., 2021b. The effects of the built environment on
quantum computing. In: 2021 IEEE International Geoscience and Remote Sensing the general health, physical activity and obesity of adults in Queensland, Australia.
Symposium IGARSS, IEEE, pp. 1571-1574. Spatial Spatio-Temporal Epidemiol. 39, 100456.
Rossetti, T., Lobel, H., Rocco, V., Hurtubia, R., 2019. Explaining subjective perceptions of Wang, C., Zhang, H., Yao, S., Yu, W., Ye, M., 2021a. MDGCN: multiple graph
public spaces as a function of the built environment: A massive data approach. convolutional network based on the differential calculation for passenger flow
Landsc. Urban Plan. 181, 169–178. forecasting in urban rail transit. J. Adv. Transport 2021, 1–10.
Rossi, A., Barlacchi, G., Bianchini, M., Lepri, B., 2019. Modelling taxi drivers’ behaviour Wang, S., Zhang, M., Huang, X., Hu, T., Li, Z., Sun, Q.C., Liu, Y., 2022b. Urban-regional
for the next destination prediction. IEEE Trans. Intell. Transp. Syst. 21, 2980–2989. disparities in mental health signals in Australia during the COVID-19 pandemic: a
Roumiani, A., Mofidi, A., 2022. Predicting ecological footprint based on global macro study via Twitter data and machine learning models. Camb. J. Reg. Econ. Soc. 15,
indicators in G-20 countries using machine learning approaches. Environ. Sci. Pollut. 663–682.
Res. 1–20. Wei, Z., Das, S., Zhang, Y., 2022. Short duration crash prediction for rural two-lane
Roumiani, A., Shayan, H., Sharifinia, Z., Moghadam, S.S., 2023. Estimation of ecological roadways: applying explainable artificial intelligence. Transp. Res. Rec. 2676,
footprint based on tourism development indicators using neural networks and 535–549.
multivariate regression. Environ. Sci. Pollut. Res. 30, 33396–33418. Werner, M., 2019. Quantum spatial computing. SIGSPATIAL Special 11, 26–33.
Saha, P., Mitra, R., Chakraborty, K., Roy, M., 2022. Application of multi layer perceptron Wirz, C.D., Xenos, M.A., Brossard, D., Scheufele, D., Chung, J.H., Massarani, L., 2018.
neural network Markov Chain model for LULC change detection in the Sub- Rethinking social amplification of risk: Social media and Zika in three languages.
Himalayan North Bengal. Remote Sens. Appl.: Soc. Environ. 26, 100730. Risk Anal. 38, 2599–2624.
Shen, H., Jiang, Y., Li, T., Cheng, Q., Zeng, C., Zhang, L., 2020. Deep learning-based air Wu, D., Gong, J., Liang, J., Sun, J., Zhang, G., 2020. Analyzing the influence of urban
temperature mapping by fusing remote sensing, station, simulation and street greening and street buildings on summertime air pollution based on street
socioeconomic data. Remote Sens. Environ. 240, 111692. view image data. ISPRS Int. J. Geo Inf. 9, 500.
Simsek, M., Kantarci, B., 2020. Artificial intelligence-empowered mobilization of Wu, H., Luo, W., Lin, A., Hao, F., Olteanu-Raimond, A.-M., Liu, L., Li, Y., 2023. SALT: A
assessments in COVID-19-like pandemics: a case study for early flattening of the multifeature ensemble learning framework for mapping urban functional zones from
curve. Int. J. Environ. Res. Public Health 17, 3437. VGI data and VHR images. Comput. Environ. Urban Syst. 100, 101921.
Siyam, N., Alqaryouti, O., Abdallah, S., 2020. Mining government tweets to identify and Xiang, L., Cai, M., Ren, C., Ng, E., 2021. Modeling pedestrian emotion in high-density
predict citizens engagement. Technol. Soc. 60, 101211. cities using visual exposure and machine learning: Tracking real-time physiology
Song, Y., Wang, R., Fernandez, J., Li, D., 2021. Investigating sense of place of the Las and psychology in Hong Kong. Build. Environ. 205, 108273.
Vegas Strip using online reviews and machine learning approaches. Landsc. Urban Xiao, Q., Liang, F., Ning, M., Zhang, Q., Bi, J., He, K., Lei, Y., Liu, Y., 2021. The long-term
Plan. 205, 103956. trend of PM2. 5-related mortality in China: The effects of source data selection.
Steane, A., 1998. Quantum computing. Rep. Prog. Phys. 61, 117. Chemosphere 263, 127894.
Stingone, J.A., Pandey, O.P., Claudio, L., Pandey, G., 2017. Using machine learning to Xie, G., Huang, L., Bin, H., Apostolidis, C., Jiang, Y., Li, G., Cai, W., 2022. Sustainable
identify air pollution exposure profiles associated with early cognitive skills among entrepreneurship in rural E-commerce: identifying entrepreneurs in practitioners by
US children. Environ. Pollut. 230, 730–740. using deep neural networks approach. Front. Environ. Sci. 370.
Suissa, O., Zhitomirsky-Geffet, M., Elmalech, A., 2022. Toward a Period-specific Xu, F., Ho, H.C., Chi, G., Wang, Z., 2019. Abandoned rural residential land: Using
optimized neural network for OCR error correction of historical hebrew texts. ACM machine learning techniques to identify rural residential land vulnerable to be
J. Comput. Cultural Heritage (JOCCH) 15, 1–20. abandoned in mountainous areas. Habitat Int. 84, 43–56.
Sun, Y., Fan, H., Bakillah, M., Zipf, A., 2015. Road-based travel recommendation using Yang, J., Ma, S., Li, Y., Zhang, Z., 2022a. Efficient data-driven crop pest identification
geo-tagged images. Comput. Environ. Urban Syst. 53, 110–122. based on edge distance-entropy for sustainable agriculture. Sustainability 14, 7825.
Sun, Q.C., Macleod, T., Both, A., Hurley, J., Butt, A., Amati, M., 2021. A human-centred Yang, K., Wu, C., Luo, Y., 2022b. The impact of COVID-19 on urban PM2. 5—taking
assessment framework to prioritise heat mitigation efforts for active travel at city Hubei Province as an example. Environ. Pollut. 294, 118633.
scale. Sci. Total Environ. 763, 143033. Yang, H., Xie, K., Ozbay, K., Ma, Y., Wang, Z., 2018. Use of deep learning to predict daily
Talukdar, S., Ghose, B., Salam, R., Mahato, S., Pham, Q.B., Linh, N.T.T., Costache, R., usage of bike sharing systems. Transp. Res. Rec. 2672, 92–102.
Avand, M., 2020. Flood susceptibility modeling in Teesta River basin, Bangladesh

16
S. Wang et al. International Journal of Applied Earth Observation and Geoinformation 128 (2024) 103734

Ye, T., Zhao, N., Yang, X., Ouyang, Z., Liu, X., Chen, Q., Hu, K., Yue, W., Qi, J., Li, Z., Zhang, H., He, J., Bao, J., Hong, Q., Shi, X., 2020a. A hybrid spatiotemporal deep
2019. Improved population mapping for China using remotely sensed and points-of- learning model for short-term metro passenger flow prediction. J Adv Transport
interest data within a random forests model. Sci. Total Environ. 658, 936–946. 2020, 1–12.
Yigitcanlar, T., Kankanamge, N., Vella, K., 2021. How are smart city concepts and Zhang, X., Zhao, X., 2022. Machine learning approach for spatial modeling of
technologies perceived and utilized? A systematic geo-Twitter analysis of smart ridesourcing demand. J. Transp. Geogr. 100, 103310.
cities in Australia. J. Urban Technol. 28, 135–154. Zhang, X., Ji, Z., Zheng, Y., Ye, X., Li, D., 2020c. Evaluating the effect of city lock-down
Yin, J., Qiu, Y., Zhang, B., 2020. Identification of poverty areas by remote sensing and on controlling COVID-19 propagation through deep learning and network science
machine learning: A case study in Guizhou, Southwest China. ISPRS Int. J. Geo Inf. models. Cities 107, 102869.
10, 11. Zhang, X., Liu, X., Chen, K., Guan, F., Luo, M., Huang, H., 2023. Inferring building
Yoo, S., Im, J., Wagner, J.E., 2012. Variable selection for hedonic model using machine function: A novel geo-aware neural network supporting building-level function
learning approaches: A case study in Onondaga County, NY. Landsc. Urban Plan. classification. Sustain. Cities Soc. 89, 104349.
107, 293–306. Zhang, Z., Wen, F., Sun, Z., Guo, X., He, T., Lee, C., 2022. Artificial intelligence-enabled
Yue, A., Mao, C., Chen, L., Liu, Z., Zhang, C., Li, Z., 2022. Detecting changes in sensing technologies in the 5G/internet of things era: from virtual reality/augmented
perceptions towards smart city on chinese social media: A text mining and sentiment reality to the digital twin. Adv. Intell Syst. 4, 2100228.
analysis. Buildings 12, 1182. Zhang, F., Zhou, B., Liu, L., Liu, Y., Fung, H.H., Lin, H., Ratti, C., 2018. Measuring human
Zeng, X., Wang, S., Zhu, Y., Xu, M., Zou, Z., 2022. A knowledge graph convolutional perceptions of a large-scale urban region using machine learning. Landsc. Urban
networks method for countryside ecological patterns recommendation by mining Plan. 180, 148–160.
geographical features. ISPRS Int. J. Geo Inf. 11, 625. Zhang, F., Wu, L., Zhu, D., Liu, Y., 2019a. Social sensing from street-level imagery: A case
Zhai, W., Bai, X., Shi, Y., Han, Y., Peng, Z.-R., Gu, C., 2019. Beyond Word2vec: An study in learning spatio-temporal urban mobility patterns. ISPRS J. Photogramm.
approach for urban functional region extraction and identification by combining Remote Sens. 153, 48–58.
Place2vec and POIs. Comput. Environ. Urban Syst. 74, 1–12. Zhang, X., Yang, Y., Zhang, Y., Zhang, Z., 2020d. Designing tourist experiences amidst air
Zhang, K., Chen, Y., Li, C., 2019b. Discovering the tourists’ behaviors and perceptions in pollution: A spatial analytical approach using social media. Ann. Tour. Res. 84,
a tourism destination by analyzing photos’ visual content with a computer deep 102999.
learning model: The case of Beijing. Tour. Manag. 75, 595–608. Zhao, X., Wang, S., Wang, H., 2022. Organizational Geosocial network: a graph machine
Zhang, J., Chen, F., Guo, Y., Li, X., 2020b. Multi-graph convolutional network for short- learning approach integrating geographic and public policy information for studying
term passenger flow forecasting in urban rail transit. IET Intel. Transport Syst. 14, the development of social organizations in China. ISPRS Int. J. Geo Inf. 11, 318.
1210–1217. Zheng, Y., Dong, C., Dong, D., Wang, S., 2021. Traffic Volume Prediction: A Fusion Deep
Zhang, Y., Cheng, T., Ren, Y., 2019c. A graph deep learning method for short-term traffic Learning Model Considering Spatial-Temporal Correlation. Sustainability 13, 10595.
forecasting on large road networks. Comput. Aided Civ. Inf. Eng. 34, 877–896. Zheng, Z., Xie, S., Dai, H.-N., Chen, X., Wang, H., 2018. Blockchain challenges and
Zhang, Y., Aslam, N.S., Lai, J., Cheng, T., 2020e. You are how you travel: A multi-task opportunities: A survey. Int. J. Web Grid Serv. 14, 352–375.
learning framework for Geodemographic inference using transit smart card data. Zhou, H., He, S., Cai, Y., Wang, M., Su, S., 2019. Social inequalities in neighborhood
Comput. Environ. Urban Syst. 83, 101517. visual walkability: Using street view imagery and deep learning technologies to
Zhang, Y., Cheng, T., Ren, Y., Xie, K., 2020f. A novel residual graph convolution deep facilitate healthy city planning. Sustaina. Cities Soc. 50, 101605.
learning model for short-term network-based traffic forecasting. Int. J. Geogr. Inf. Zhu, R., Lin, D., Jendryke, M., Zuo, C., Ding, L., Meng, L., 2018. Geo-tagged social media
Sci. 34, 969–995. data-based analytical approach for perceiving impacts of social events. ISPRS Int. J.
Zhang, Y., Chen, Z., Zheng, X., Chen, N., Wang, Y., 2021. Extracting the location of Geo Inf. 8, 15.
flooding events in urban systems and analyzing the semantic risk using social sensing Zhu, D., Zhang, F., Wang, S., Wang, Y., Cheng, X., Huang, Z., Liu, Y., 2020.
data. J. Hydrol. 603, 127053. Understanding place characteristics in geographic contexts through graph
convolutional neural networks. Ann. Am. Assoc. Geogr. 110, 408–420.

17