arXiv:2312.08827v1  [cs.AI]  14 Dec 2023
Artiﬁcial Intelligence and Human Geography
Chapter in the Encyclopedia of Human Geography
Song Gao
Geospatial Data Science Lab, Department of Geography, University of
Wisconsin-Madison, 550 N Park Stret, Madison, 53706, WI, USA.
Contributing authors: song.gao@wisc.edu;
Abstract
This paper examines the recent advances and applications of AI in human geogra-
phy especially the use of machine (deep) learning, including place representation
and modeling, spatial analysis and predictive mapping, and urban planning
and design. AI technologies have enabled deeper insights into complex human-
environment interactions, contributing to more eﬀective scientiﬁc exploration,
understanding of social dynamics, and spatial decision-making. Furthermore,
human geography oﬀers crucial contributions to AI, particularly in context-aware
model development, human-centered design, biases and ethical considerations,
and data privacy. The synergy beween AI and human geography is essential
for addressing global challenges like disaster resilience, poverty, and equitable
resource access. This interdisciplinary collaboration between AI and geography
will help advance the development of GeoAI and promise a better and sustainable
world for all.
Keywords: AI, GeoAI, human-in-the-loop, ethics, trustworthiness
1 Introduction
Artiﬁcial intelligence (AI) is a ﬁeld in computer science and engineering that focuses
on developing intelligent machines capable of performing problem-solving tasks and
achieving goals that typically require human intelligence (Turing, 1950; McCarthy,
2004). There are generally two types of AI: Weak AI and Strong AI. Weak AI sits
at the foundation of AI development, with its focused human-like speciﬁc abilities
(e.g., conversation in Siri) and practical applications (e.g., facial recognition from
images and videos) in speciﬁc contexts, while Strong AI (known as Artiﬁcial General
1

Intelligence) represents the ultimate goal of AI research, aiming to replicate or even
surpass human intelligence to solve problems, learn, and plan for the future with
self-aware consciousness (Goertzel, 2014).
AI is revolutionizing many domains including geographical sciences, presenting
both immense opportunities and formidable challenges. The rapid advancement of AI
is fueled by theoretical breakthroughs in neural science and computer science, the
ubiquity of big data, advanced computer hardware such as graphics processing units
(GPUs) and powerful high-performance computing platforms that enable the eﬃcient
development training, and deployment of AI models, and the ﬂourishing AI products,
services and applications that are changing human behaviors (Gao, 2020; Li, 2020;
Torrens, 2018).
As Couclelis (1986) expressed her excitement about the AI-driven evolution—“It
is not often that geography is touched by a development having the potential to aﬀect
substantially all of the practical, technical, methodological, theoretical and philo-
sophical aspects of our work.” The intersection of AI and geography has a rich and
evolving history, with its early roots documented in works by Couclelis (1986); Smith
(1984); Openshaw and Openshaw (1997). Prior to the surge of deep learning research
in the 2010s (LeCun et al, 2015) and its applications in geography and earth sci-
ences (Hu et al, 2019b; Goodchild and Li, 2021; Liu and Biljecki, 2022; Reichstein et al,
2019), signiﬁcant AI advancements included theoretical explorations in the 1950s and
1960s (Buchanan, 2005); the emergence of artiﬁcial neural networks, heuristic search
algorithms, knowledge-based symbolic expert systems, neurocomputing, and artiﬁcial
life concepts (e.g., cellular automata) in the 1980s; the development of genetic pro-
gramming, fuzzy logics, and hybrid intelligent systems in the 1990s (Openshaw and
Openshaw, 1997); the integration of ontology and semantics for information retrieval
and knowledge graphs in the 2000s; and the recent development of foundation model
(Bommasani et al, 2021; Mai et al, 2023). All of these advances have laid the founda-
tion for the burgeoning ﬁeld of Geospatial artiﬁcial intelligence (GeoAI) (Gao et al,
2023).
Rooted in geography and geographic information science (GIScience), GeoAI
has now emerged as a transformative interdisciplinary ﬁeld, integrating geographi-
cal studies with AI techniques, particularly spatially-explicit machine learning and
deep learning methods as well as geo-knowledge graphs (Gao, 2020; Janowicz et al,
2020; Mai et al, 2022). This convergence has spurred groundbreaking advancements
in both academia and industry. GeoAI encompasses the development of intelligent
computer programs that emulate human perception, spatial reasoning, and the ability
to uncover insights from geographical phenomena and understanding their complex
dynamics through coupling spatial knowledge and earth science process-based mod-
els with deep neural networks (Gao, 2021; Reichstein et al, 2019). The researches in
GeoAI can deepen our understanding of complex human-environmental systems and
their interactions, particularly within a spatial context.
Regarding the topics of AI in human geography, a recent conversation about GeoAI,
counter-AI, and human geography by Janowicz et al (2022) discussed opportunities
and challenges in deﬁning intelligence for machines, and the role of humans in AI
2

development and decision-making processes, emphasizing the need for ethical consid-
erations and accountability in the development of GeoAI. In the following, I will ﬁrst
introduce some of the recent advances of AI in human geography and then discuss the
role of human geography in AI development.
2 AI in Human Geography
Human geography has a long tradition of studying the spatially and temporally vary-
ing interrelationships between people, place, and environment, and the ways in which
people and their activities shape, and are shaped by, the physical, economic, political,
and cultural environment (Jones, 2012). The integration of AI technologies has signif-
icantly expanded the capacity of researchers and practitioners in understanding the
complex human-environment interactions. Geographers have explored a wide range of
topics and made advances in the following areas (but not limited to).
Place Representation and Modeling: Place can serve as a function between
location and people (Mennis and Mason, 2016), a function of location, activity and
time (McKenzie and Adams, 2017), and a function of social relations (Giordano and
Cole, 2018). Traditionally, data about places were collected through mapping agencies
(e.g., digital gazetteers) and survey-based narratives. The emergence of geospatial big
data brings new opportunities to extract ﬁne spatiotemporal resolution of human-place
interaction data and understand the rich place semantics from large-scale volunteered
geographic information and crowdsourced data streams, such as social media posts
(including texts, photos, and videos), GPS tracks, text reviews on points of interest
(POIs) and neighborhoods, and other Web documents (Gao et al, 2017b,c; Liu et al,
2015). A wide-range of AI methods provide new opportunities to understand and
extract the characteristics of places as well as associated human activities, experiences,
emotions, and movements in diﬀerent contexts. For example, urban areas of interest
that attract people’s attention but with diﬀerent spatial patterns and region-speciﬁc
semantics were automatically extracted from georeferenced photos posted on social
media by employing spatial and spectral clustering algorithms, and natural language
processing techniques (Hu et al, 2015). In addition, the spatial and hierarchical seman-
tics between places that support human cognition of places, identiﬁcation of urban
functional regions, and information retrieval in digital maps, were modelled using POI
embeddings (i.e., multidimensional vectors) via topic modeling (Gao et al, 2017a)
and deep learning techniques (Huang et al, 2022). Patterns and relations between
places can be computed and extracted from collective human descriptions using place
graphs (Chen et al, 2018). Empowered by the state-of-the-art human face and emo-
tion recognition AI techniques, human emotions at diﬀerent places were extracted
from millions of georeferenced photos (Kang et al, 2019), which enabled the study of
spatially embodied emotions in human geography (Simonsen et al, 2007). Recently,
researchers found that geo-knowledge-guided large language models (e.g., ChatGPT)
improved the extraction of location textual descriptions from disaster-related social
media messages, which can facilitate collaborations between disaster response experts
and AI developers and ultimately help save human lives (Hu et al, 2023).
3

Spatial Analysis and Predictive Mapping: AI has revolutionized geospatial
data analysis, interpretation, and modeling of human-environment relations by pro-
viding innovative spatial analytical tools and methodologies. New or modiﬁcation of
spatial analysis methods and spatial statistical models have been proposed by incor-
porating deep neural networks, which we termed as Intelligent Spatial Analytics (Zhu
et al, 2022a). Geographically weighted artiﬁcial neural network (GWANN) (Hage-
nauer and Helbich, 2022) and geographically neural network weighted regression
(GWNNR) (Du et al, 2020) models were developed by constructing the spatial non-
stationary weights with neural networks and feeding into the geographically weighted
regression for more accurate predictive modeling. Similarly, new spatial regression
graph convolutional neural networks (SRGCNNs) were developed for the spatial
regression analysis of geographical multivariate distributions and outperformed tra-
ditional spatial autoregressive models (Zhu et al, 2022b). Machine learning and deep
learning algorithms have also been widely used for data-driven modeling and pre-
dictive mapping in geography (Miller and Goodchild, 2015) and identifying patterns
and relationships that would be nearly impossible for human researchers to discern
manually (Lavallin and Downs, 2021). For example, Zhu et al (2020) applied the
graph convolutional neural networks (GCNNs) for modeling place graphs and infer-
ring the unknown properties of a place with a high prediction accuracy by utilizing
both the observed attributes (e.g., visual and functional features) and the relational
characteristics (e.g., distance, adjacency, spatial interaction intensity) of the places.
When applying state-of-the-art deep learning models for population mapping in geog-
raphy, the issues on model selection, neighboring eﬀects, and systematic biases play an
important role on the predictive mapping accuracy (Huang et al, 2021). Researchers
found that population estimation performance reduced with increasing neighborhood
sizes and a pervasive bias existed in which all tested deep learning models (i.e.,
VGG, ResNet, Xception, and DenseNet) overestimated sparsely populated patches and
underestimated densely populated ones (Huang et al, 2021). This work also demon-
strated the importance of spatial concepts and some limitations when applying AI
methods in human geography due to spatial heterogeneity. Future eﬀorts need to be
made towards the (weak) replication of GeoAI models across space and time in the
social and environmental sciences (Goodchild and Li, 2021).
Urban Planning and Design:
Human geographers have long been interested
in understanding the intricate relationships between people and the environment and
collaborating with urban planners for planning our cities. By applying AI algorithms,
they can eﬀectively interpret large geospatial datasets, including census surveys, satel-
lite imagery, street-view images, social media posts, and other sensor data to support
urban analytics (De Sabbata et al, 2023). By utilizing the state-of-the-art deep learning
models, human perception of neighborhood playability for childhood development in
cities (Kruse et al, 2021) and urban visual intelligence about the hidden neighborhood
socioeconomic status, such as poverty status and health outcomes, can be extracted
from street-view images (Fan et al, 2023; Kang et al, 2020); near real-time global land
cover and land use patterns and temporal changes can be detected from 10m spatial
resolution of Sentinel-2 remote sensing imagery (Brown et al, 2022). Urban geographers
4

and planners can also extract valuable insights into population dynamics, traﬃc pat-
terns, land use patterns, environmental changes, and the interactions between social
and ecological systems, so as to optimize urban spatial structure, improving trans-
portation eﬃciency and quality of living, and environmental sustainability (Mortaheb
and Jankowski, 2023; Liu and Biljecki, 2022). AI-driven predictive modeling can assist
in rooftop solar potential estimation for sustainable city design (Wu and Biljecki,
2021), and anticipating changes in population density and mobility, facilitating better
resource allocation and disaster preparedness in (Zou et al, 2022; Vongkusolkit and
Huang, 2021). Additionally, AI-powered semantic and sentiment analyses of online
neighborhood reviews helped better understand the perceptions and emotions of peo-
ple toward their living neighborhoods and environments, which can be applied for
supporting urban planning and improving quality of life studies (Hu et al, 2019a).
Furthermore, the advances in generative AI models brings exiting opportunities for
urban planners to facilitate the automatic rendering of urban master plans (Ye et al,
2022) and building ﬂoorplan layouts (Wu et al, 2022) via generative adversarial net-
works (GANs), otherwise they need to rely on subjective design and labor-intensive
production process. However, the generative AI also introduces the deep fakes into
geography (Zhao et al, 2021) and we need to make more eﬀorts on the investigation
of trustworthiness of such AI generated geospatial data and city plans.
3 Human Geography for AI
Human geography also contributes to the development and applications of AI by
providing valuable insights into the social, cultural, and economic factors that shape
human behavior and interactions with the technology.
Context-Aware and Debiasing AI:
Human geography contributes to AI
through the development of context-aware models and the analysis of biases of AI
models when applied in heterogeneous contexts and places. Human perception of the
physical and socioeconomic environments through deep learning has attracted large
interest of scholars from geography, computer sciences, urban planning, and environ-
mental psychology (Gebru et al, 2017; Fan et al, 2023; Kang et al, 2020; Han et al,
2022). Human perception of places reﬂects people’s subjective feelings towards their
surrounding environment, e.g., whether a place is safe, lively, and wealthy (Zhang
et al, 2018). However, there exist multiple types of biases in existing AI-driven compu-
tational frameworks when quantifying human perceptions of places: data bias, model
bias, and perception bias. Kang et al (2023) compared the safety perception mea-
sures from the survey based on neighborhood residents’ responses with those from
the GeoAI approach with street-view imagery to better understand the relationship
between the two types of measures. This research found that citywide residents, but
not neighborhood residents, may feel economically vibrant places look safe; elder peo-
ple may underestimate the safety of their living places which may enlarge perception
bias. Therefore, integrating cultural, historical, and socioeconomic insights into pre-
trained AI models and ﬁne-tuning them with local data and participant inputs may
enhance AI’s ability to comprehend and respond to spatial heterogeneity and diverse
5

human experiences. This approach not only improves the accuracy of AI applications
but also fosters inclusivity and respect for social diversity across geographic regions.
Data Privacy and Ethical Framework:
Human geography addresses ethi-
cal concerns related to data collection and usage, as well as community trust (Rowe,
2021). This knowledge is invaluable for AI developers who must navigate the ethi-
cal implications of collecting and analyzing vast amounts of data regarding privacy
and security. In GeoAI development, such concerns often refer to the use or exposure
of sensitive geospatial information such as a user’s home location, workspace, POI
preferences, daily trajectories, and associated personal inferences based on such infor-
mation (Keßler and McKenzie, 2018; Rao et al, 2023b). There is a trade-oﬀbetween
the data utility in downstream tasks (e.g., user proﬁling) and user privacy (Gao et al,
2019). In human mobility studies, common user privacy protection methods include
de-identiﬁcation, geomasking, trajectory k-anonymization, and diﬀerential privacy.
Recently, deep neural networks such as LSTM and GANs have been used in human
movement trajectory generation tasks with the aim to balance the data utility-privacy
trade-oﬀ(Rao et al, 2020, 2023b). With the substantial advances in AI foundation
models, recent studies, however, revealed that the development and use of founda-
tion models could potentially unveil substantial privacy and security risks, including
the disclosure of sensitive information, representational bias, hallucinations, and mis-
use (Bommasani et al, 2021). In the lifecycle of building and using GeoAI foundation
models (Mai et al, 2023), a series of potential privacy and security risks that existed
around the pre-training and ﬁne-tuning stages with geospatial data, centralized serv-
ing and tooling, prompting-based interaction, and feedback mechanisms (Rao et al,
2023a). Looking ahead, human geographers and GIScience scholars can further con-
tribute to establishing guidelines that ensure responsible and privacy-conscious GeoAI
practices.
Human-Centered Design: Understanding the perspectives, experiences, emo-
tions, and lived realities of people in diﬀerent places and regions is essential for human
geography research (Tuan, 1979). Therefore, human geography emphasizes the impor-
tance of human-centered design in place-based innovation and policy making, aligning
seamlessly with the principles of ethical (Geo-)AI development (Siau and Wang, 2020;
Kang et al, 2024) and user-centered design in cartography (Roth, 2017). By incorpo-
rating human geographers in the (Geo-)AI design processes, developers can create AI
systems that better serve diverse communities and mitigate potential negative impacts
on our society, which is well aligned with the core idea of keeping “human-in-the-loop”
for the development of AI systems and algorithms (Janowicz et al, 2022; Zanzotto,
2019). Several principles and practices like collaborative engagement with data (Lloyd
and Dykes, 2011), accountability mechanisms in AI systems design (Mittelstadt, 2019)
and empathy (Srinivasan and Gonz´alez, 2022) can be considered for addressing the
social-technical challenges in the AI paradigm.
4 Conclusion
The relationship between AI and human geography is not a one-way street. As
AI becomes increasingly sophisticated, its applications in human geography will
6

become even more diverse and impactful. At the same time, the abovementioned
critical insights provided by human geographers will be crucial for ensuring that
less-biased (Geo-)AI is developed and deployed in a responsible and ethical man-
ner. Mitigating multiple types of biases throughout the “data-model-action” loop in
human-environment systems may ensure equitable outcomes and AI-driven spatial
decision making. By working together, AI and human geography can address some
of the most pressing challenges facing human-environment nexus today, including cli-
mate change, poverty, food security, health, disaster resilience, and equitable access to
resources and services. This interdisciplinary collaboration has the potential to create
a better and sustainable world for all.
References
Bommasani R, Hudson DA, Adeli E, et al (2021) On the opportunities and risks of
foundation models. arXiv preprint arXiv:210807258
Brown CF, Brumby SP, Guzder-Williams B, et al (2022) Dynamic world, near real-
time global 10 m land use land cover mapping. Scientiﬁc Data 9(1):251
Buchanan BG (2005) A (very) brief history of artiﬁcial intelligence. AI Magazine
26(4):53–53
Chen H, Winter S, Vasardani M (2018) Georeferencing places from collective human
descriptions using place graphs. Journal of Spatial Information Science (17):31–62
Couclelis H (1986) Artiﬁcial intelligence in geography: Conjectures on the shape of
things to come. The Professional Geographer 38(1):1–11
De Sabbata S, Ballatore A, Miller HJ, et al (2023) Geoai in urban analytics
Du Z, Wang Z, Wu S, et al (2020) Geographically neural network weighted regres-
sion for the accurate estimation of spatial non-stationarity. International Journal of
Geographical Information Science 34(7):1353–1377
Fan Z, Zhang F, Loo BP, et al (2023) Urban visual intelligence: Uncovering hidden city
proﬁles with street view images. Proceedings of the National Academy of Sciences
120(27):e2220417120
Gao S (2020) A review of recent researches and reﬂections on geospatial artiﬁcial
intelligence. Geomatics and Information Science of Wuhan University 45(12):1865–
1874
Gao S (2021) Geospatial artiﬁcial intelligence (GeoAI). Oxford University Press
Gao S, Janowicz K, Couclelis H (2017a) Extracting urban functional regions from
points of interest and human activities on location-based social networks. Transac-
tions in GIS 21(3):446–467
7

Gao S, Janowicz K, Montello DR, et al (2017b) A data-synthesis-driven method
for detecting and extracting vague cognitive regions. International Journal of
Geographical Information Science 31(6):1245–1271
Gao S, Li L, Li W, et al (2017c) Constructing gazetteers from volunteered big geo-data
based on hadoop. Computers, Environment and Urban Systems 61:172–186
Gao S, Rao J, Liu X, et al (2019) Exploring the eﬀectiveness of geomasking techniques
for protecting the geoprivacy of twitter users. Journal of Spatial Information Science
(19):105–129
Gao S, Hu Y, Li W (2023) Handbook of Geospatial Artiﬁcial Intelligence. CRC Press,
URL https://doi.org/10.1201/9781003308423
Gebru T, Krause J, Wang Y, et al (2017) Using deep learning and google street view
to estimate the demographic makeup of neighborhoods across the united states.
Proceedings of the National Academy of Sciences 114(50):13108–13113
Giordano A, Cole T (2018) The limits of gis: Towards a gis of place. Transactions in
GIS 22(3):664–676
Goertzel B (2014) Artiﬁcial general intelligence: concept, state of the art, and future
prospects. Journal of Artiﬁcial Intelligence 5(1):1
Goodchild MF, Li W (2021) Replication across space and time must be weak in the
social and environmental sciences. Proceedings of the National Academy of Sciences
118(35):e2015759118
Hagenauer J, Helbich M (2022) A geographically weighted artiﬁcial neural network.
International Journal of Geographical Information Science 36(2):215–235
Han X, Wang L, Seo SH, et al (2022) Measuring perceived psychological stress in
urban built environments using google street view and deep learning. Frontiers in
Public Health 10:891736
Hu Y, Gao S, Janowicz K, et al (2015) Extracting and understanding urban areas
of interest using geotagged photos. Computers, Environment and Urban Systems
54:240–254
Hu Y, Deng C, Zhou Z (2019a) A semantic and sentiment analysis on online neigh-
borhood reviews for understanding the perceptions of people toward their living
environments. Annals of the American Association of Geographers 109(4):1052–1073
Hu Y, Li W, Wright D, et al (2019b) Artiﬁcial intelligence approaches. The Geographic
Information Science & Technology Body of Knowledge
Hu Y, Mai G, Cundy C, et al (2023) Geo-knowledge-guided gpt models improve
the extraction of location descriptions from disaster-related social media messages.
8

International Journal of Geographical Information Science 37(11):2289–2318
Huang W, Cui L, Chen M, et al (2022) Estimating urban functional distributions
with semantics preserved poi embedding. International Journal of Geographical
Information Science 36(10):1905–1930
Huang X, Zhu D, Zhang F, et al (2021) Sensing population distribution from satel-
lite imagery via deep learning: Model selection, neighboring eﬀects, and systematic
biases. IEEE Journal of Selected Topics in Applied Earth Observations and Remote
Sensing 14:5137–5151
Janowicz K, Gao S, McKenzie G, et al (2020) GeoAI: spatially explicit artiﬁcial intel-
ligence techniques for geographic knowledge discovery and beyond. International
Journal of Geographical Information Science 34(4):625–636
Janowicz K, Sieber R, Crampton J (2022) Geoai, counter-ai, and human geography:
A conversation. Dialogues in Human Geography 12(3):446–458
Jones A (2012) Human geography: The basics. Routledge
Kang Y, Jia Q, Gao S, et al (2019) Extracting human emotions at diﬀerent places
based on facial expressions and spatial clustering analysis. Transactions in GIS
23(3):450–480
Kang Y, Zhang F, Gao S, et al (2020) A review of urban physical environment sensing
using street view imagery in public health studies. Annals of GIS 26(3):263–275
Kang Y, Abraham J, Ceccato V, et al (2023) Assessing diﬀerences in safety perceptions
using geoai and survey across neighbourhoods in stockholm, sweden. Landscape and
Urban Planning 236:104768
Kang Y, Gao S, Roth R (2024) Artiﬁcial intelligence studies in cartography: A review
and synthesis of methods, applications, and ethics. pp 1–20
Keßler C, McKenzie G (2018) A geoprivacy manifesto. Transactions in GIS 22(1):3–19
Kruse J, Kang Y, Liu YN, et al (2021) Places for play: Understanding human percep-
tion of playability in cities using street view images and deep learning. Computers,
Environment and Urban Systems 90:101693
Lavallin A, Downs JA (2021) Machine learning in geography–past, present, and future.
Geography Compass 15(5):e12563
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. nature 521(7553):436–444
Li W (2020) GeoAI: Where machine learning and big data converge in GIScience.
Journal of Spatial Information Science (20):71–77
9

Liu P, Biljecki F (2022) A review of spatially-explicit geoai applications in urban
geography. International Journal of Applied Earth Observation and Geoinformation
112:102936
Liu Y, Liu X, Gao S, et al (2015) Social sensing: A new approach to understanding our
socioeconomic environments. Annals of the Association of American Geographers
105(3):512–530
Lloyd D, Dykes J (2011) Human-centered approaches in geovisualization design: Inves-
tigating multiple methods through a long-term case study. IEEE Transactions on
Visualization and Computer Graphics 17(12):2498–2507
Mai G, Hu Y, Gao S, et al (2022) Symbolic and subsymbolic geoai: Geospatial
knowledge graphs and spatially explicit machine learning
Mai G, Huang W, Sun J, et al (2023) On the opportunities and challenges of foundation
models for geospatial artiﬁcial intelligence. arXiv preprint arXiv:230406798 pp 1–32
McCarthy J (2004) What is artiﬁcial intelligence? Stanford University pp 1–15
McKenzie G, Adams B (2017) Juxtaposing thematic regions derived from spatial and
platial user-generated content. In: 13th international conference on spatial informa-
tion theory (COSIT 2017), Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik
Mennis J, Mason MJ (2016) Modeling place as a relationship between a person and a
location. In: International Conference on GIScience Short Paper Proceedings
Miller HJ, Goodchild MF (2015) Data-driven geography. GeoJournal 80:449–461
Mittelstadt B (2019) Principles alone cannot guarantee ethical ai. Nature Machine
Intelligence 1(11):501–507
Mortaheb R, Jankowski P (2023) Smart city re-imagined: City planning and geoai in
the age of big data. Journal of Urban Management 12(1):4–15
Openshaw S, Openshaw C (1997) Artiﬁcial intelligence in geography. John Wiley &
Sons, Inc.
Rao J, Gao S, Kang Y, et al (2020) Lstm-trajgan: A deep learning approach to tra-
jectory privacy protection. In: 11th International Conference on Geographic Infor-
mation Science (GIScience 2021), Schloss Dagstuhl-Leibniz-Zentrum f¨ur Informatik,
p 12
Rao J, Gao S, Mai G, et al (2023a) Building privacy-preserving and secure geospatial
artiﬁcial intelligence foundation models. In: The 31st ACM International Conference
on Advances in Geographic Information Systems (SIGSPATIAL’23). ACM, pp 1–4
10

Rao J, Gao S, Zhu S (2023b) Cats: Conditional adversarial trajectory synthesis
for privacy-preserving trajectory data publication using deep learning approaches.
International Journal of Geographical Information Science 37(12):2538–2574
Reichstein M, Camps-Valls G, Stevens B, et al (2019) Deep learning and process
understanding for data-driven earth system science. Nature 566(7743):195–204
Roth R (2017) User interface and user experience (ui/ux) design. Geographic
Information Science & Technology Body of Knowledge 2017(Q2):1–10
Rowe F (2021) Big data and human geography. SocArXiv Papers 2021(0):1–10
Siau K, Wang W (2020) Artiﬁcial intelligence (ai) ethics: ethics of ai and ethical ai.
Journal of Database Management (JDM) 31(2):74–87
Simonsen K, et al (2007) Practice, spatiality and embodied emotions: An outline of a
geography of practice. Human Aﬀairs (2):168–181
Smith TR (1984) Artiﬁcial intelligence and its applicability to geographical problem
solving. The Professional Geographer 36(2):147–158
Srinivasan R, Gonz´alez BSM (2022) The role of empathy for artiﬁcial intelligence
accountability. Journal of Responsible Technology 9:100021
Torrens PM (2018) Artiﬁcial intelligence and behavioral geography. Handbook of
Behavioral and Cognitive Geography 20(0):357–371
Tuan YF (1979) Space and place: humanistic perspective. In: Philosophy in geography.
Springer, p 387–427
Turing AM (1950) Computing machinery and intelligence. Mind 49(0):433–460
Vongkusolkit J, Huang Q (2021) Situational awareness extraction: a comprehensive
review of social media data classiﬁcation during natural hazards. Annals of GIS
27(1):5–28
Wu AN, Biljecki F (2021) Roofpedia: Automatic mapping of green and solar roofs for
an open roofscape registry and evaluation of urban sustainability. Landscape and
Urban Planning 214:104167
Wu AN, Stouﬀs R, Biljecki F (2022) Generative adversarial networks in the built
environment: A comprehensive review of the application of gans across data types
and scales. Building and Environment p 109477
Ye X, Du J, Ye Y (2022) Masterplangan: Facilitating the smart rendering of urban
master plans via generative adversarial networks. Environment and Planning B:
Urban Analytics and City Science 49(3):794–814
11

Zanzotto FM (2019) Human-in-the-loop artiﬁcial intelligence. Journal of Artiﬁcial
Intelligence Research 64:243–252
Zhang F, Zhou B, Liu L, et al (2018) Measuring human perceptions of a large-scale
urban region using machine learning. Landscape and Urban Planning 180:148–160
Zhao B, Zhang S, Xu C, et al (2021) Deep fake geography? when geospatial data
encounter artiﬁcial intelligence. Cartography and Geographic Information Science
48(4):338–352
Zhu D, Zhang F, Wang S, et al (2020) Understanding place characteristics in geo-
graphic contexts through graph convolutional neural networks. Annals of the
American Association of Geographers 110(2):408–420
Zhu D, Gao S, Cao G (2022a) Towards the intelligent era of spatial analysis and
modeling. In: Proceedings of the 5th ACM SIGSPATIAL International Workshop
on AI for Geographic Knowledge Discovery. ACM, pp 10–13
Zhu D, Liu Y, Yao X, et al (2022b) Spatial regression graph convolutional neu-
ral networks: A deep learning paradigm for spatial multivariate distributions.
GeoInformatica 26(0):645—-676
Zou L, Cai H, Mandal D, et al (2022) Empowering smart disaster response with social
media and geoai. In: AGU Fall Meeting, vol 2022. American Geophysical Union, pp
NH15C–0329
12
