About the symposium
Spatial and temporal thinking is important not just because everything happens at some places and at some time, but because knowing where and when things are happening is key to understanding how and why they happened or will happen. Spatial data science is concerned with the representation, modeling, and simulation of spatial processes, as well as with the publication, retrieval, reuse, integration, and analysis of such space- and place-centric data. It generalizes and unifies research from fields such as geographic information science/geoinformatics, geo/spatial statistics, remote sensing, environmental studies, and transportation studies, and fosters applications of methods developed in these fields in other disciplines ranging from social to biological and physical sciences.
Data-driven methods, such as machine learning models, have been attracting attention from the Geoscience community for the past several years. For instance, they have been successfully used to quantify the semantics of place types, to classify geo-tagged images, to predict traffic and air quality, to improve resolution of remotely sensed images, to recognize objects in such imagery, to predict and compare trajectories, to name but a few. Geospatial observations may be vague, uncertain, heterogeneous, dependent on other nearby observations, biased, and multimodal; thus, spatial and temporal principles should be included in data science techniques such as deep neural networks. Unsurprisingly, research has shown that by doing so, we can substantially outcompete more general (non-spatial) models when applied to geo-data or applications with a spatial and temporal component.
To keep this discussion alive and help the community to exchange ideas and lessons learned about spatial and temporal aspects of data science, we are hosting the 5th Spatial Data Science Symposium (SDSS 2024) as a distributed virtual meeting. The symposium aims to bring together researchers from both academia and industry to discuss experiences, insights, methodologies, and applications, taking spatial and temporal knowledge into account while addressing their domain-specific problems. The format of this symposium will be a combination of keynotes, scientific sessions, as well as paper presentations. In contrast to classical conferences, the community will decide on those sessions, and the main focus will be on interaction. Hence, we welcome submissions for both papers and sessions (see below). SDSS 2024 will be a distributed symposium in a sense that while the event as such will be online, we will host (and help others to host) individual get-togethers to jointly experience the symposium in person.
DATES
Session submission deadline: August 5 August 16, 2024
Paper submission deadline: August 30 September 9, 2024
Paper notification: September 30, 2024
Symposium Dates: October 23-24, 2024
Keynote Speakers
This year we are happy to welcome two keynote speakers who are working on the cutting edge of spatial data science.
Clio Andris
Associate Professor of City & Regional Planning and Interactive ComputingGeorgia Institute of Technology
Social Networks and SNoMaN: A Visual Analytic Tool for Spatial Social Network Mapping and Analysis
Abstract
Spatial social networks (SSNs) are node-link structures that provide evidence interpersonal or inter-organizational relationships, where nodes and edges have a defined geographic location. To model spatial social networks, users need both geographic and social network metrics. Following the research framework of Exploratory Spatial Data Analysis (ESDA) and design principles of social network analysis tools, we will share design goals of exploratory spatial social network analysis (SSNA). Guided by these design goals, this talk will introduce SNoMaN (Social Network Mapping and Analysis), a visual analytic tool which links network and geographical layouts and helps users conduct SSNA by interactively computing and visualizing SSN metrics, describing spatial distributions, exploring associations, and detecting anomalies. We will also show new types of visual diagrams, including Cluster-Cluster Plots, Centralization Plots, on-the-fly mapping of geometrically-bounded network modules, and Route Factor Diagrams. The ultimate goal is to help non-geographers learn more about their social networks and to help geographers and spatial data scientists integrate social network concepts into their flow data analyses
Biography
Clio Andris is an associate professor in the School of City and Regional Planning and the School of Interactive Computing at Georgia Tech. She directs the Friendly Cities Lab and conducts research on geographic information science, social networks, spatial networks, and geovisualization. Her lab is a member of the Information Visualization Lab in the School of Interactive Computing and is affiliated with the Institute for People and Technology (IPaT). She received a PhD from MIT in 2011 in Urban Information Systems where she was an NDSEG fellow and member of the Senseable City Lab. She held postdoctoral positions at the Singapore-MIT Alliance for Research and Technology and at the Santa Fe Institute. Prior to Georgia Tech, she was a faculty member in the Department of Geography at Penn State, and her lab was an affiliate of the GeoVISTA Center. She won an NSF CAREER award in 2021 and serves on the National Geospatial Advisory Committee.
Bernd Resch
Professor for Geosocial Artificial IntelligenceIT:U Interdisciplinary Transformation University
Data, Ideation and AI: On the Role of Data Characteristics, Hypotheses and Context in Geospatial Research
Abstract
The growing influence of digital data and tools, along with recent advancements in AI, is transforming how we approach research. With the ever increasing availability of large, open datasets—ranging from social media posts and news articles to physiological measurements and mobile phone records—entirely new fields of study are emerging. This, in turn, fundamentally changes our geospatial research paradigms. Nowadays, hypothesis-driven and data-driven approaches are no longer diametrical, but even more so depend on each other. In fact, the analysis of new large data sources coupled with stringent hypothesis-building has the potential to revolutionise geospatial research through mutually informing each other. Additionally, new ways of contextualising analysis results through leveraging the multimodal nature of currently available data creates entirely new opportunities for generating new insights into geospatial processes.
Biography
Bernd Resch is a Full Professor for Geo-social AI at IT:U and a Visiting Scholar at Harvard University (USA). His research interest revolves around understanding cities as complex systems through analysing a variety of digital data sources, focusing on developing machine learning algorithms for analysing human-generated data like social media posts and physiological measurements from wearable sensors. The findings are relevant to a number of fields including urban research, disaster management, epidemiology, and others.
Live Interview
The symposium will feature a live/on-stage, interactive interview with Matt Duckham from RMIT University.
Matt Duckham
Professor in Geospatial SciencesRMIT University
Matt Duckham is the Director of the Information in Society EIP (Enabling Impact Platform) and Professor in Geospatial Sciences at RMIT University. At RMIT, he has occupied a number of senior leadership roles, including: Acting Dean STEMM Diversity and Inclusion, Associate Dean of Geospatial Science, and Deputy Head of the School of Mathematical and Geospatial Sciences. Previously, he was also Science Director for the CRC for Spatial Information (CRCSI) Rapid Spatial Analytics Program and before joining RMIT he was Professor in Geographic Information Science within the department of Infrastructure Engineering at the University of Melbourne, when he also held a visiting Professor position and the University of Greenwich.
Prof. Duckham's research focuses on the area of Geographic Information Science, particularly distributed and robust computation and visualisation with uncertain spatial and spatiotemporal information, within the domain of mobile, location aware and sensor enabled systems.
Early Career Panel
Sara Fabrikant
University of ZurichMeiliu Wu
University of GlasgowSimon Scheider
Utrecht UniversityKathleen Stewart
University of MarylandProgram
Wednesday, October 23
CEST | EDT | Session |
---|---|---|
13:00 | 7:00 | Opening and Interview (60 mins) |
14:00 | 8:00 | Break |
14:15 | 8:15 | Keynote: Clio Andris - Georgia Tech University (60 mins) |
15:15 | 9:15 | Break |
15:30 | 9:30 | Thematic Session: Multimodal Social Sensing: Towards a Holistic Understanding of Geospatial and Social Process (75 mins) |
16:45 | 10:45 | Break |
17:00 | 11:00 | Paper Session: Presentation of accepted peer-reviewed papers (60 mins) |
18:00 | 12:00 | Break (60 mins) |
19:00 | 13:00 | Thematic Session: Geoprivacy Challenges and Solutions in the Digital Society (75 mins) |
20:15 | 14:15 | Break |
20:30 | 14:30 | Thematic Session: Spatial Representation Learning and Location Encoding (75 mins) |
Thursday, October 24
CEST | EDT | Session |
---|---|---|
13:00 | 7:00 | Keynote: Bernd Resch - Interdisciplinary Transformation University (60 mins) |
14:00 | 8:00 | Break |
14:15 | 8:15 | Thematic Session: Urban Analytics: Leveraging Big Data and GeoAI for Sustainable Cities (75 mins) |
15:30 | 9:30 | Break |
15:45 | 9:45 | Early Career Session: Panel (60 mins) |
16:45: | 10:45 | Break |
17:00 | 11:00 | Thematic Session: Teaching reproducibility and replicability in spatial data science: Where to start and what to do. (75 mins) |
18:15 | 12:15 | Break (60 mins) |
19:15 | 13:15 | Thematic Session: Spatial Data Insights in Urban Accessibility (75 mins) |
20:30 | 14:30 | Closing |
Accepted Papers
These papers were peer-reviewed by the program committee and accepted for publication and presentation at the symposium.
Standardizing Machine Learning APIs for Earth Observation Data Cubes
Brian Pondi and Edzer Pebesma
Exploring Changes in Sense of Place Post-Events Using Large Language Models
Mina Karimi, Ivan Majic, and Krzysztof Janowicz
Understanding LLMs' capabilities to support spatially-disaggregated epidemiological simulations
Anuj Goenka, Ilya Zaslavsky, Jiaxi Lei, and Rishi Graham
Thematic Sessions
Along with regular paper presentations, this year's symposium will feature a series of thematic sessions organized by a variety of teams from around the globe. The session titles and organizers are listed below. More details will be added as they become available.
Geoprivacy Challenges and Solutions in the Digital Society
As digital mapping and location-based services become integral to daily life, the issue of geoprivacy grows increasingly critical. This panel will explore the intersection of privacy, technology, and geography, focusing on the ethical and technological challenges associated with geospatial data collection and usage. Participants will gain insights from experts on public perceptions of geoprivacy and innovative solutions on geoprivacy protections.
More details
Hongyu Zhang
University of Massachusetts AmherstEun-Kyeong Kim
Luxembourg Institute of Socio-Economic ResearchTeaching reproducibility and replicability in spatial data science-Where to start and what to do
A consensus is emerging across the sciences that the reproducibility of research must be improved and that more independent reproductions and replications are needed to assess the state of existing knowledge. A similar view is emerging among GIScientists and spatial data scientists. Multiple authors have called for improvements in the transparency and accessibility of procedures and data, the adoption of open science practices, and the execution of reproduction and replication studies. However, a parallel consensus has yet to develop about how to teach reproducibility in spatial data science. Educators have no competency model to use to structure their curricula, no set of best practices to inform their class exercises, and few model courses to emulate at their own institutions.
More details
Peter Kedron
University of California, Santa BarbaraJoseph
Holler
MiddleburyCollege
Andrew Trgovac
Arizona State UniversityUrban Analytics: Leveraging Big Data and GeoAI for Sustainable Cities
As cities continue to expand and evolve, the need for data-driven decision-making has never been more critical. This session will explore how big data can be effectively utilized to gain insights into urban dynamics, and examine the role of GeoAI in analyzing vast amounts of geospatial data, enabling urban planners and policymakers to predict, visualize, and optimize various aspects of urban life such as health, transportation, and safety.
More details
Shenjun Yao
East China Normal UniversityJue Wang
University of TorontoSpatial Data Insights in Urban Accessibility
To build truly inclusive cities and societies, we must leverage cutting-edge spatial data science methodologies to uncover and address hidden accessibility barriers. This special session at SDSS 2024 explores innovative spatial data-driven techniques and mapping projects reshaping our understanding of urban accessibility.The session will feature the Mapping Our Cities for All (MOCA) project, which tackles urban inaccessibility by crowdsourcing building accessibility information through participatory mapping and the lived experiences of people with disabilities. In partnership with AccessNow, we will examine how MOCA data reveals insights into building accessibility and uses spatial methods such as machine learning, geocoding, and double geocoding to predict industry codes (NAICS) for businesses.
More details
Victoria Fast
University of CalgaryShiloh Deitz
Saint Luis UniversityAchilleas Psyllidis
TU DelftVasileios Milias
TU DelftRoos Teeuwen
TU DelftBlake Walker
FAU Erlangen-NürnbergMultimodal Social Sensing: Towards a Holistic Understanding of Geospatial and Social Process
Social sensing as an established field of research has traditionally focused on analysing social media posts, mobile phone records, or other types of user-generated data. Most research efforts have performed unimodal analysis, i.e., deriving patterns from a single type of data characteristic. This session explores new ways of analysing multiple modalities of different data sources in joint, machine learning approaches, focusing on spatially explicit ML, in order to generate deeper insights into geospatial and social processes.
More details
Bernd Resch
Interdisciplinary Transformation UniversityShaily Gandhi
University of SalzburgDavid Hanny
Interdisciplinary Transformation University AustriaSpatial Representation Learning and Location Encoding
We will be hosting a tutorial on TorchSpatial, a cutting-edge learning framework and benchmark for spatial representation learning (SRL). SRL is essential for learning general-purpose neural network representations from various types of spatial data (e.g., points, polylines, polygons, networks, images, etc.) in their native formats. Learning good spatial representations is fundamental for various downstream applications such as species distribution modeling, weather forecasting, trajectory generation, population estimation, sustainability prediction, geographic question answering, etc.
More details
Gengchen Mai
University of Texas at AustinNi
Lao
GoogleXiaobai Angela Yao
University of GeorgiaNemin
Wu
University of GeorgiaQian
Cao
University of Georgia
Zhangyu Wang
UC Santa BarbaraCall for Papers
We welcome short papers (3,000 words) and vision papers (2,000 words) on the following (or similar) topics:
- Geospatial thinking in the arts
- Spatial and temporal knowledge representation and reasoning
- Geospatial artificial intelligence (GeoAI) & spatially explicit machine learning
- Neuro-symbolic representation learning for spatial and temporal data
- Spatial and temporal data mining
- Spatial and spatiotemporal data uncertainty
- Geographic information retrieval
- Geospatial knowledge graphs
- Geospatial semantics
- Spatial statistics / Geostatistics
- Geo-simulation
- Diversity, inclusion, and equity in spatial data science
- Social and environmental ethics in spatial data science
- Geospatial applications that use data-driven methods, including but not limited to:
- Movement analysis
- Disaster response
- Environmental studies
- Geoprivacy
- Social sensing
- Location-based services
- Humanitarian relief
- Crime analysis
- Urban analytics
Submission Guidelines
All submissions must be original and must not be simultaneously submitted to another journal or conference/workshop. All submissions must be in English and formatted according to the LNCS template (Overleaf template here). Proceedings of the symposium will be published as open access SDSS proceedings through ArXiv.org. Each accepted paper will be assigned an individual DOI. All submissions will be peer-reviewed by the Program Committee.
Organizing Committee
Bruno
Martins
Program Chair
University of Lisbon
Grant
McKenzie
Program Chair
McGill University
Judith
Verstegen
Program Chair
Utrecht University
Shenjun
Yao
Program Chair
East China Normal University
Rui
Zhu
Program Chair
University of Bristol
Krzysztof Janowicz
General ChairUniversity of Vienna
Program Committee
- Fernando Benitez-Paez, University of St Andrews
- Vanessa Brum-Bastos, University of Canterbury
- Victoria Fast, University of Calgary
- Vanessa Frias-Martinez, University of Maryland
- Anita Graser, Austrian Institute of Technology
- Yingjie Hu, University at Buffalo
- Steffen Knoblauch, Heidelberg University
- Ourania Kounadi, University of Vienna
- Ross Purves, University of Zurich
- Johannes Scholz, Paris Lodron University of Salzburg
- Tuomas Väisänen, University of Helsinki
- Shaohua Wang, International Research Center of Big Data for Sustainable Development Goals, China
- Hongyu Zhang, University of Massachusetts, Amherst
- Qunshan Zhao, University of Glasgow
- More TBA...
Symposium Hubs
SDSS2024 is a distributed/online symposium. Participants are welcome to join one of the symposium hubs distributed around the world. Groups of participants will meet at these hubs to present and discuss with other participants both in person and online.
If you are interested in hosting a hub in your city, please email grant.mckenzie@mcgill.ca.
Vienna, Austria
University of ViennaContact: krzysztof.janowicz@univie.ac.at
Montreal, Canada
McGill UniversityContact: grant.mckenzie@mcgill.ca
Bristol, UK
University of BristolContact:
rui.zhu@bristol.ac.uk
Shanghai, China
East China Normal UniversityContact:
sjyao@geo.ecnu.edu.cn
Boston, USA
University of Massachusetts, Amherst (Mt. Ida Campus)Contact:
honzhang@umass.edu
Heidelberg, Germany
Heidelberg UniversityContact:
steffen.knoblauch@uni-heidelberg.de
Calgary, Canada
University of CalgaryContact: victoria.fast@ucalgary.ca
Saint Louis, USA
Saint Louis UniversityContact:
shiloh.deitz@slu.edu
Erlangen, Germany
University of Erlangen-NurembergContact:
blake.walker@fau.de