The paper led by my student M. Naser Lessani, “SGWR: similarity and geographically weighted regression”, became the #1 most read articles published last year in the International Journal of Geographical Information Science (IJGIS). Read it here: https://lnkd.in/eUHrkPND

The paper led by professor Siqin (Sisi) Wang from USC, “GPT, large language models (LLMs) and generative artificial intelligence (GAI) models in geospatial science: a systematic review”, became the #1 most read articles published last year in the International Journal of Digital Earth (IJDE). Read it here: https://lnkd.in/e8pk9xeA

Our autonomous GIS definition paper, co-authored with my student Huan Ning, titled “Autonomous GIS: the next-generation AI-powered GIS”, has became the #1 most read and top 5 most cited articles of the past three years, and is among the top 10 most read articles of all time in IJDE! Read it here: https://lnkd.in/eibR34pb

We invite you to explore those works, and hope you find them interesting and inspiring!

We are thrilled to announce our latest vision paper (preprint), “GIScience in the Era of Artificial Intelligence: A Research Agenda Towards Autonomous GIS,” led by Dr. Zhenlong Li and Ph.D. candidate Huan Ning from the Geoinformation and Big Data Research Lab at Penn State. This paper, a collaborative effort involving GIScience experts across academia, national labs, and government agencies, presents a timely research agenda for the next evolutionary stage of GIS—one that is autonomous, intelligent, more accessible, and powered by AI.

This work is rooted in the lab’s ongoing innovative research on integrating AI into spatial analysis. Over the past two years, we have been developing foundational ideas and prototype systems that explore how large language models (LLMs) can serve as decision-making cores and build geoprocessing workflows in GIS applications. This vision paper builds on our earlier work, particularly the 2023 paper that formally introduced and defined the concept of Autonomous GIS (Li & Ning, 2023), by expanding the conceptual framework, demonstrates early-stage implementations, and outlines a roadmap for advancing this emerging paradigm….. Read the full post here.

You can access the preprint here and the original Autonomous GIS definition paper here.

If you are attending AAG this year, we cordially invite you to join our 13 in-person sessions focused on geospatial big data, spatial computing, autonomous GIS, human mobility, disaster management, and public health!

GIBD lab members recently participated in the 6th National Big Data Health Science Conference, held on February 13-14, 2025, at the Pastides Alumni Center, University of South Carolina, Columbia. The conference, themed “Unlocking the Power of Big Data in Health: Transforming Data into Actionable Intelligence,” brought together experts to discuss advancements in health data science.

GIBD Lab members contributed significantly to the event:

• Huan Ning delivered an oral presentation on “A Mapping Toolkit for Built Environment Auditing Based on Street View Imagery”

• Naser Lessani presented a poster titled “Leveraging Large Language Models for Systematic Reviewing: A Case Study Using HIV Medication Adherence Research”.

• Temitope Akinboyewa presented a poster titled “Smartphone-Based Place Visitation Data Better Explain Neighborhood-Level Coronary Heart Disease in the United States”.

Additionally, the lab hosted an exhibition table, demonstrating our recent research activities to attendees.

The 14th ISDE International Lectures, themed “Harnessing the Power of Generative AI in GIScience”, successfully took place on 21 January 2025, attracting over 3,400 participants worldwide via platforms such as Zoom and the ISDE Bilibili channel. This event brought together leading experts in the field of GIScience to explore the diverse applications of Generative AI in understanding human-environment interactions, disaster management, and the development of next-generation GIS systems. Dr. Zhenlong Li presented on “Autonomous GIS: The Next-Generation AI-Powered GIS.” He introduced the concept of the Autonomous GIS, which leverages advanced AI technologies to create intelligent and adaptive geospatial systems. Dr. Li showcased a prototype called LLM-Geo that uses large language models to automate spatial data collection, analysis, and visualisation. The system demonstrated its ability to retrieve human mobility data and visualise trends, highlighting the potential of Autonomous GIS to revolutionise spatial analysis and decision-making processes. In addition to presenting, Dr. Li also served as the event’s moderator, facilitating engaging discussions throughout the session.

• Presentation slides: https://www.digitalearth-isde.org/uploadfile/2025/0124/20250124110315312.pdf
• More information: https://www.digitalearth-isde.org/show-34-46-1.html

We are pleased to announce that GIBD Lab will be hosting a series of very interesting in-person sessions at the 2025 AAG Annual Meeting in Detroit!  We warmly invite you to join us and present your work in one of our sessions if it aligns with your research. To participate, simply email your abstract code to one of the organizers listed in the session. Please include the name of the session you’d like to present in when submitting your abstract. Let’s come together to share insights and advance research in geospatial big data. We look forward to your contributions!

1. Harnessing Geospatial Big Data for Infectious Diseases
2. Geospatial Big Data for Analyzing and Understanding Human Mobility Patterns
3. Social Sensing and Big Data Computing for Disaster Management
4. Big Data Computing for Geospatial Applications
5. Big Data Computing for Geospatial Applications (2)
6. Harnessing the Power of Generative AI in GIScience through Autonomous GIS Agents
7. Integrative Approaches to Understanding Human Mobility and Health Outcomes
8. Urban Sensing and Understanding via Geospatial Big Data and AI (I)
9. Urban Sensing and Understanding via Geospatial Big Data and AI (II)
10. GIScience and Hazards (I)
11. GIScience and Hazards (II)
12. Uncertainties in Big Data Analytics in Disaster Research
13. GeoAI and Deep Learning Symposium: Generative AI in GIScience: a research agenda towards Autonomous GIS (Panel Session)

Presentations

1. GIS Copilot: Towards an Autonomous GIS Agent for Spatial Analysis
2. An anti-Asian area racism index
3. Understanding Urban Park Visitors’ Pattern in New York City and the Growing Inequity Issues: A longitudinal Study with SafeGraph Foot Traffic Data
4. Enhancing computational efficiency of the similarity and geographically weighted (SGWR) regression model and introducing its Python implementation
5. LLM-Find: An Autonomous GIS Agent Framework for Geospatial Data Retrieval and Downloading
6. Autonomous GIS: the next-generation AI-powered GIS
7. Comparative Analysis of Human Mobility Patterns: Utilizing Taxi and Mobile (SafeGraph) Data to Investigate Neighborhood-Scale Mobility in New York City 

The Geoinformation and Big Data Research Lab have unveiled an innovative tool, GIS Copilot, which integrates Large Language Models (LLMs) into Geographic Information Systems (GIS) to enable users to perform spatial analysis using natural language. The GIS Copilot is a step closer to achieving the broader vision of Autonomous GIS, which aim at democratizing access to spatial analysis, making it accessible to users of all expertise levels.

GIS Copilot operates as a plugin for the QGIS platform, enabling users to perform spatial operations through simple natural language, spanning from basic operations such as

• Can you please create 2000-feet zones around each health facilities in Washington DC to identify areas of service coverage?
• Generate contour lines from the DEM of Puerto Rico with a 50-meter interval.

to more complex questions such as

• Generate an obesity risk behavior index of each county in the contiguous US by analyzing the rate of visits to unhealthy food retailers (such as convenience store, alcoholic drinking places, and limited service restaurant) and the visit rate to places that support physical activity (e.g., sports centers, parks, fitness centers). Visualize the results in a thematic map to highlight the obesity risk behavior index across counties.
• Could you analyze and visualize the fast food accessibility score for each county based on the number of fast food restaurants and population using a thematic map with blue graduated colors. Then, analyze the correlation between the county-level obesity rate and the fast food accessibility score by drawing a scatter plot with a regression line.

to the development of interactive web mapping applications such as

• Generate an interactive web map using leaflet for the shown data layer.

The tool’s functionality was rigorously evaluated on more than 100 spatial analysis tasks, categorized into three levels of complexity including Basic Tasks: Single-step operations involving one GIS tool and data layer. Intermediate Tasks: Multi-step processes requiring multiple tools and user guidance. Advanced Tasks: Complex, multi-step analyses where the tool independently determines and executes workflows without explicit user input. Results showed that GIS Copilot excels at automating basic and intermediate tasks, with significant progress in handling advanced workflows. While challenges remain in achieving complete autonomy for highly complex tasks, the tool represents a major step toward the vision of autonomous GIS.

The release of GIS Copilot has sparked massive interest in the geospatial and AI communities. A LinkedIn post announcing the tool has garnered  over 150,000 impressions, 2,200 likes, and more than 220 reposts within just a few days. The overwhelming response reflects the demand for such an GIS Copilot that simplify GIS workflows and enhance accessibility.

GIS Copilot’s source code is available on GitHub, with the plugin downloadable from the official QGIS plugin page. The research team has also made data and case studies used in testing accessible online, inviting collaboration and feedback from the global GIS community.

The GIS Copilot represents a significant milestone in the development of Autonomou GIS by integrating AI with GIS, bridging the gap between technical GIS expertise and practical application. This innovation not only simplifies geospatial workflows but also enhances decision-making across diverse domains such as disaster management, urban planning, and public health.

To learn more about GIS Copilot’s design, implementation, and discussions, please check out our preprint paper.

Geographic information system (GIS) users and analyst need to fetching geospatial data for analysis or research tasks. Data fetching can be time-consuming and label intensive. Our recent study proposes LLM-Find, an autonomous GIS agent framework to retrieve geospatial data by generating and executing programs with self-debugging. LLM-Find adopts an LLM as the decision maker to pick up the applicable data source from a list and then fetch data from the selected source. Each data source has a pre-defined handbook that records the metadata and technical details for data fetching. The proposed framework is flexible and extensible, designed as a plug-and-play mechanism; human users or autonomous data scrawlers can add a new data source by adding a new handbook. LLM-Find provides a fundamental agent framework for data fetching in autonomous GIS. We also prototyped an agent based on LLM-Find, which can fetch data from OpenStreetMap, COVID-19 cumulative cases from GitHub, administrative boundaries and demographic data from the US Census Bureau, weather data from a commercial provider, satellite basemap from ESRI World Imagery, and worldwid DEM from OpenTopography.org.

We tested various data cases; by accepting data requests in natural language, most of the requests got correct data in an about 80% – 90% success rate. We feel excited about that because the success of such data fetching agent indicates that the data intensive GIS research or boarder scientific research can be executed by agents. Autonomous research agents can collect necessary online or local data and then conduce analysis parallely while adjust methods or strategies for better results. LLM-Find will be a foundational role in such a bright vision.

For more details, please refer to our paper: Ning, Huan, Zhenlong Li, Temitope Akinboyewa, and M. Naser Lessani. 2024. “An Autonomous GIS Agent Framework for Geospatial Data Retrieval” arXiv. https://doi.org/10.48550/arXiv.2407.21024. GitHub repository: https://github.com/gladcolor/LLM-Find 

Further reading: Autonomous GIS: the next-generation AI-powered GIS. Recommended citation format: Li Z., Ning H., 2023. Autonomous GIS: the next-generation AI-powered GIS. International Journal of Digital Earth. GitHub repository: https://github.com/gladcolor/LLM-Geo

Our new article titled “Automated floodwater depth estimation using large multimodal model for rapid flood mapping” is published in Computational Urban Science. 

Read full article here!

Abstract

Information on the depth of floodwater is crucial for rapid mapping of areas affected by floods. However, previous approaches for estimating floodwater depth, including field surveys, remote sensing, and machine learning techniques, can be time-consuming and resource-intensive. This paper presents an automated and rapid approach for estimating floodwater depth from on-site flood photos. A pre-trained large multimodal model, Generative pre-trained transformers (GPT-4) Vision, was used specifically for estimating floodwater. The input data were flood photos that contained referenced objects, such as street signs, cars, people, and buildings. Using the heights of the common objects as references, the model returned the floodwater depth as the output. Results show that the proposed approach can rapidly provide a consistent and reliable estimation of floodwater depth from flood photos. Such rapid estimation is transformative in flood inundation mapping and assessing the severity of the flood in near-real time, which is essential for effective flood response strategies.

Keywords: Flood mapping, Large multimodal model, Large language model, ChatGPT, GeoAI, Disaster management

Akinboyewa, T., Ning, H., Lessani, M.N. et al. Automated floodwater depth estimation using large multimodal model for rapid flood mapping. Comput.Urban Sci. 4, 12 (2024). https://doi.org/10.1007/s43762-024-00123-3

Our new research paper “SGWR: similarity and geographically weighted regression” is published in the International Journal of Geographical Information Science (IJGIS). In this study, we extend the geographically weighted regression (GWR) by integrating attribute similarity alongside the conventional geographically weighted matrix. The new model, called SGWR, was evaluated across various datasets, including housing prices, crime rates, and three health outcomes including mental health, depression, and HIV. Results show that SGWR consistently outperforms the global regression model and the traditional GWR based on several statistical measures across all experimental datasets.

Read the full article (open access) at https://lnkd.in/enejjskZ
Code and datasets are available at https://lnkd.in/e4Yn6xyP
Graphic user interface (GUI) for SGWR is under development. Stay tuned!

Our new article led by Dr. Manzhu Yu, titled “Assessing the 2023 Canadian wildfire smoke impact in Northeastern US: Air quality, exposure and environmental justice”, is published in the Science of The Total Environment. 

Abstract: The Canadian wildfires in June 2023 significantly impacted the northeastern United States, particularly in terms of worsened air pollution and environmental justice concerns. While advancements have been made in low-cost sensor deployments and satellite observations of atmospheric composition, integrating dynamic human mobility with wildfire PM2.5 exposure to fully understand the environmental justice implications remains under investigated. This study aims to enhance the accuracy of estimating ground-level fine particulate matter (PM2.5) concentrations by fusing chemical transport model outputs with empirical observations, estimating exposures using human mobility data, and evaluating the impact of environmental justice. Employing a novel data fusion technique, the study combines the Weather Research and Forecasting model with Chemistry (WRF-Chem) outputs and surface PM2.5 measurements, providing a more accurate estimation of PM2.5 distribution. The study addresses the gap in traditional exposure assessments by incorporating human mobility data and further investigates the spatial correlation of PM2.5 levels with various environmental and demographic factors from the US Environmental Protection Agency (EPA) Environmental Justice Screening and Mapping Tool (EJScreen). Results reveal that despite reduced mobility during high PM2.5 levels from wildfire smoke, exposure for both residents and individuals on the move remains high. Regions already burdened with high environmental pollution levels face amplified PM2.5 effects from wildfire smoke. Furthermore, we observed mixed correlations between PM2.5 concentrations and various demographic and socioeconomic factors, indicating complex exposure patterns across communities. Urban areas, in particular, experience persistent high exposure, while significant correlations in rural areas with EJScreen factors highlight the unique vulnerabilities of these populations to smoke exposure. These results advocate for a comprehensive approach to environmental health that leverages advanced models, integrates human mobility data, and addresses socio-demographic disparities, contributing to the development of equitable strategies against the growing threat of wildfires.

Read the full article here.

Our new paper titled “An MPI-based parallel genetic algorithm for multiple geographical feature label placement based on the hybrid of fixed-sliding models” is accepted by Geo-spatial Information Science.

Abstract: Multiple geographical feature label placement (MGFLP) has been a fundamental problem in geographic information visualization for decades. Moreover, the nature of label positioning has proven to be an NP-hard problem. Although advances in computer technology and robust approaches have addressed the problem of label positioning, the lengthy running time of MGFLP has not been a major focus of recent studies. Based on a hybrid of the fixed-position and sliding models, a Message Passing Interface (MPI) parallel genetic algorithm is proposed in the present study for MGFLP to label mixed types of geographical features. To evaluate the quality of label placement, a quality function is defined based on four quality metrics: label-feature conflict; label-label conflict; label association with the corresponding feature; label position priority for all three types of features. The experimental results show that the proposed algorithm outperforms the DDEGA, DDEGA-NM, and Parallel-MS in both label placement quality and computation time efficiency. Across three datasets, compared to Parallel-MS, running times decreased from 118.45 to 8.34, 45.98 to 3.51, and 20.01 to 0.43 min, with further reductions in label-label and label-feature conflicts.

Keywords: Label placement; fixed position; geographical features; parallel genetic algorithm; Message Passing Interface

To Cite: Lessani M.N., Li Z., Deng J., Guo Z., (2024). An MPI-based parallel genetic algorithm for multiple geographical feature label placement based on the hybrid of fixed-sliding models, Geo-spatial Information Science. https://doi.org/10.1080/10095020.2024.2313326

Our study on the bias of mobile location data (SafeGraph or Advan) is published in PLOS ONE, a peer-reviewed journal. The smartphone-based human mobility data has been widely used in social research, but still lacked a comprehensive bias investigation for such datasets. Our study provides an understanding and assessment framework for mobility data. Please check out the paper and feel free to contact us for any questions!

Understanding the bias of mobile location data across spatial scales and over time: a comprehensive analysis of SafeGraph data in the United States

Abstract: Mobile location data has emerged as a valuable data source for studying human mobility patterns in various contexts, including virus spreading, urban planning, and hazard evacuation. However, these data are often anonymized overviews derived from a panel of traced mobile devices, and the representativeness of these panels is not well documented. Without a clear understanding of the data representativeness, the interpretations of research based on mobile location data may be questionable. This article presents a comprehensive examination of the potential biases associated with mobile location data using SafeGraph Patterns data in the United States as a case study. The research rigorously scrutinizes and documents the bias from multiple dimensions, including spatial, temporal, urbanization, demographic, and socioeconomic, over a five-year period from 2018 to 2022 across diverse geographic levels, including state, county, census tract, and census block group. Our analysis of the SafeGraph Patterns dataset revealed an average sampling rate of 7.5% with notable temporal dynamics, geographic disparities, and urban-rural differences. The number of sampled devices was strongly correlated with the census population at the county level over the five years for both urban (r > 0.97) and rural counties (r > 0.91), but less so at the census tract and block group levels. We observed minor sampling biases among groups such as gender, age, and moderate-income, with biases typically ranging from -0.05 to +0.05. However, minority groups such as Hispanic populations, low-income households, and individuals with low levels of education generally exhibited higher levels of underrepresentation bias that varied over space, time, urbanization, and across geographic levels. These findings provide important insights for future studies that utilize SafeGraph data or other mobile location datasets, highlighting the need to thoroughly evaluate the spatiotemporal dynamics of the bias across spatial scales when employing such data sources.

Keywords: population bias, human mobility, smartphone data, Advan Patterns, SafeGraph

Paper link:

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0294430

GIBD’s recent work, LLM-Geo, was introduced in the 2023 annual conference of GIS Society of Japan. LLM-Geo is a prototype of autonomous GIS that can conduct spatial analysis without human intervention. About 50 GIS professionals tried LLM-Geo in a hands-on session, following a brief introductory presentation by lab member Huan Ning, a PhD student at Penn State. Hands-on participants showed interest in LLM-Geo’s analysis and plotting capability and felt excited about its potential advancements. The GIBD team will continue to improve LLM-Geo. https://www.tandfonline.com/doi/full/10.1080/17538947.2023.2278895

Dr. Chaowei Phil Yang, Professor at George Mason University and the Founding Director of the NSF Spatiotemporal Innovation Center, will be presenting a seminar at the University of South Carolina Big Data Health Science Center, sharing insights on “Utilizing Big Spatiotemporal Data to Understand COVID and its Impacts”. The seminar is scheduled for Wednesday, August 30, at 10:30 AM, Discovery Building RM 140.

Check out our new research proposing Autonomous GIS as the next-generation AI-powered GIS.

Link the full preprint: https://www.researchgate.net/publication/370635187_Autonomous_GIS_the_next-generation_AI-powered_GIS

Code for LLM-Geo: https://github.com/gladcolor/LLM-Geo 

Autonomous GIS: the next-generation AI-powered GIS

Abstract: Large Language Models (LLMs), such as ChatGPT, demonstrate a strong understanding of human natural language and have been explored and applied in various fields, including reasoning, creative writing, code generation, translation, and information retrieval. By adopting LLM as the reasoning core, we introduce Autonomous GIS (AutoGIS) as an AI-powered geographic information system (GIS) that leverages the LLM’s general abilities in natural language understanding, reasoning and coding for addressing spatial problems with automatic spatial data collection, analysis and visualization. We envision that autonomous GIS will need to achieve five autonomous goals including self-generating, self-organizing, self-verifying, self-executing, and self-growing. We developed a prototype system called LLM-Geo using the GPT-4 API in a Python environment, demonstrating what an autonomous GIS looks like and how it delivers expected results without human intervention using two case studies. For both case studies, LLM-Geo returned accurate results, including aggregated numbers, graphs, and maps, significantly reducing manual operation time. Although still lacking several important modules such as logging and code testing, LLM-Geo demonstrates a potential path towards next-generation AI-powered GIS. We advocate for the GIScience community to dedicate more effort to the research and development of autonomous GIS, making spatial analysis easier, faster, and more accessible to a broader audience.

Keywords: Autonomous Agent, GIS, Artificial Intelligence, Spatial Analysis, Large Language Models, ChatGPT

Figure 1. Overall workflow of LLM-Geo

Results automatically generated by LLM-Geo for counting the population living near hazardous wastes. (a) Solution graph, (b) assembly program (Python codes), and (c) returned population count and generated map.

We are pleased to share that GIBD alumnus Daniel Newsome has taken on a new role as a Data Analytics Manager with Rise Interactive! Congratulations, Daniel!

Dr. Zhenlong Li (Institutional PI) and Dr. Grace Yan from Missouri S&T (PI) receive $190,000 funding from Taylor Geospatial Institute for a new project titled “Developing a SMART Geospatial Tool to Classify Building Archetype at Community Level towards Digital-Twins of Disaster Resilience (DTDR)“.

Project Goal: To achieve intergenerational, equitable resilience under the changing climate (society sustainability), considering the uncertainties in projecting future natural hazards, the overarching goal of this project is to develop powerful, evolving Digital-Twins for Disaster Resilience (DTDR), with cutting-edge GeoSpatial tools and engineering and social science breakthrough discoveries integrated. The DTDR will create future potential climate/hazard scenarios and embed potential engineering hardening approaches and nature-based solutions for hazard mitigation. It will allow stakeholders to experiment with hazard mitigation solutions and practice evidence-based decision making on resilience planning and implementation. This will eventually lead to intergenerational, equitable resilience policy for both physical/social subsystems, empowering communities to confront current and future natural hazards.

Our new article entitled “Revealing geographic transmission pattern of COVID-19 using neighborhood-level simulation with human mobility data and SEIR model: A case study of South Carolina” led by PhD Candidate Huan Ning is published in the International Journal of Applied Earth Observation and Geoinformation (IF: 7.672) by Elsevier.  The article is freely available at https://www.sciencedirect.com/science/article/pii/S1569843223000687

Abstract: Direct human physical contact accelerates COVID-19 transmission. Smartphone mobility data has emerged as a valuable data source for revealing fine-grained human mobility, which can be used to estimate the intensity of physical contact surrounding different locations. Our study applied smartphone mobility data to simulate the second wave spreading of COVID-19 in January 2021 in three major metropolitan statistical areas (Columbia, Greenville, and Charleston) in South Carolina, United States. Based on the simulation, the number of historical county-level COVID-19 cases was allocated to neighborhoods (Census block groups) and points of interest (POIs), and the transmission rate of each allocated place was estimated. The result reveals that the COVID-19 infections during the study period mainly occurred in neighborhoods (86%), and the number is approximately proportional to the neighborhood’s population. Restaurants and elementary and secondary schools contributed more COVID-19 infections than other POI categories. The simulation results for the coastal tourism Charleston area show high transmission rates in POIs related to travel and leisure activities. The results suggest that neighborhood-level infectious controlling measures are critical in reducing COVID-19 infections. We also found that households of lower socioeconomic status may be an umbrella against infection due to fewer visits to places such as malls and restaurants associated with their low financial status. Control measures should be tailored to different geographic locations since transmission rates and infection counts of POI categories vary among metropolitan areas.

South Carolina has a shortage of providers, and many children must be placed away from their original communities. This makes it difficult for birth parents, for children, for visitation and for providing health care. This project aims to produce statistics and maps on the distances from the children and youth original “home address” to their current foster care “provider” address and the distances from the child to the county office and from the provider to the county office for the South Carolina Department of Social Services (SCDSS). While SCDSS understands that many children and youth are placed outside of their county of origin, this project will provide additional insights and specificity.

Huan Ning, our PhD candidate, is selected to receive one of the 14 Breakthrough Graduate Scholar awards across the entire USC system. Big congratulations, Huan!

https://www.sc.edu/about/offices_and_divisions/research/news_and_pubs/news/2023/2023_Breakthrough_Awards_Announcement.php

“Each year, the Office of the Vice President for Research receives a large number of nominations for each of the Breakthrough awards, and a committee of faculty reviewers selects award recipients from the many remarkable faculty and graduate student nominees representing the entire USC system. The close competition embodies the spirit of strong, high-level research conducted here at the university. This year, we are very pleased to present two Breakthrough Leadership in Research awards, 12 Breakthrough Star awards and 14 Breakthrough Graduate Scholar awards to our 2023 class of Breakthrough awardees.

Vice President for Research Julius Fridriksson praised this notable group, saying, ‘As the research enterprise continues to grow at USC so does the quality of our researchers. This year’s awardees are a true testament to the accomplishments and quality of mentorship, research and the next generation of researchers here at USC. It’s an honor to be a considered a colleague of among the 2023 Breakthrough Awardees.‘ ”

The two datasets (cellphone-based population flow data and Twitter data) are among the four research datasets released by the USC Big Data Health Science Center as part of the Research Data Repositories (RDR) initiative.

Check out the Twitter data description at https://bigdata.sc.edu/twitter-data/

Check out the cellphone-based population flow (ODT flow) data description at https://bigdata.sc.edu/cell-phone-based-place-visitation-data/

News release: https://bigdata.sc.edu/bdhsc-rdr-announcement/

Alexander Fulham successfully defended his thesis proposal next Tuesday, Dec. 20 at 12:30 PM in Callcott 228. His proposal title is “Sentiment analysis of Swedish and Finnish Twitter users’ views toward NATO pre- and post- 2022 Russian reinvasion of Ukraine“. His committee includes Drs. Zhenlong Li, Carl Dahlman, and Robert Kopack. Congratulations, Alex!

The Big Data Health Science Center is seeking research and program-based abstracts for oral and poster presentations responsive to its 4^th annual conference theme, “Unlocking the Power of Big Data in Health: Translating Data Science into Program Development and Implementation.” This conference will be held in-person at the Pastides Alumni Center in Columbia, SC on February 10-11, 2023. Accepted abstracts will be published in conference proceedings.

Research and program-based abstracts in the areas of infectious diseases, other health conditions, artificial intelligence methods for sensing and diagnosis, electronic health records, social media research, genomics analysis, geospatial sciences, infrastructural and capacity development, professional development, community/industry engagement, academic training, and methodological advances are desired.

Learn more about abstract submissions and current 2023 keynote speakers at www.sc-bdhs-conference.org

Check out our new article titled “Exploring large-scale spatial distribution of fear of crime by integrating small sample surveys and massive street view images“, published in Environment and Planning B Urban Analytics and City Science.

The COVID-19 pandemic has imposed catastrophic impacts on the restaurant industry as a crucial socioeconomic sector that contributes immensely to the global economy. However, what remains incomplete is our quantitative understanding of how the restaurant industry was recovered from COVID-19 in terms of restaurant visitations and revenue, customers’ origins as well as the relationship between restaurant visitations and travel distances. Existing studies in the context of COVID-19 mainly reply on survey data and cannot reveal the changing spectrum of the restaurant industry at a large spatial and temporal scale. Here we construct a spatially explicit evaluation of the effect of COVID-19 on the restaurant industry in the United States, drawing on the attributes of +200,000 restaurants from Yelp and +600 million individual-level restaurant visitations provided by SafeGraph from 1 January 2019 to 31 December 2021. We produce quantitative evidence of lost restaurant visitations and revenue amid the COVID-19 pandemic, the changes in the areal characteristics of customers’ origins, and the retained visitation law of human mobility-the number of restaurant visitations decreases as the inverse square of their travel distances-though such a distance-decay effect varies across metropolitan areas and becomes marginal at the later stage of the pandemic. Our findings support policy makers to monitor economic relief and design place-based policies for economic recovery.

Read the full preprint article at:

https://www.researchgate.net/publication/363794683_Diverged_landscape_of_restaurant_recovery_the_effect_of_COVID-19_on_the_restaurant_industry_in_the_United_States

Check out the news release about our recent publication about the COVID-19 impact on the black-owned restaurants, led by Dr. Xiao Huang, our lab alumnus and now the Assistant Professor at the Department of Geosciences, University of Arkansas.

https://www.eurekalert.org/news-releases/963549

Full article: https://www.researchgate.net/publication/361780077_Black_businesses_matter_A_longitudinal_study_of_black-owned_restaurants_in_the_COVID-19_pandemic_using_geospatial_big_data

GIBD receives $30,000 from the USC BDHSC Pilot Project Program to conduct a pilot study of developing a novel network-based big data approach to measure healthcare utilization disparity. The project team include Drs. Zhenlong Li, Shan Qiao, Bankole Olatosi, and Jiajia Zhang.

Project summary: Healthcare utilization is a critical factor that influences population health and wellbeing. To identify, explain, and address disparities and inequities in healthcare utilization, it is necessary to develop a valid measurement approach that can accurately capture the disparities and explore the factors that contribute to the disparities in a timely manner. Increasing attention is being paid to developing constructs or measurement approaches that can reflect complex interplays of factors at multiple socioecological levels. The availability of healthcare Big Data (e.g., large place visitation data sampled from mobile devices and electronic health records [EHR]) and advanced Big Data analytics makes it possible to use Big Data approaches to address existing knowledge gaps in measurement methodology including a lack of real-world evidence, limited availability of real-time and large-coverage datasets, and a dearth of studies applying multilevel perspectives. In this pilot project, we propose to develop a network-based big data approach to measure and visualize disparities in healthcare utilization in South Carolina (SC). Specifically, we will first develop a machine learning-based network prediction model to construct a statewide healthcare visitation network using cellphone-based place visitation data and ground-truth EHR data (for model training and refining); based on the validated statewide healthcare visitation network, we will detect actual catchment areas of healthcare facilities and develop healthcare utilization measures (indices) using geographically constrained network partition and aggregation. We will then test the performance and utility of the network-based big data approach in revealing healthcare utilization patterns using multivariate geo-visualization. Leveraging our fruitful collaboration with the state’s health department and health agencies and successful experiences with implementing NIH-funded Big Data studies since 2017, we will be able to develop this network-based big data approach for analyzing healthcare utilization disparity, which, with proven efficacy, will contribute to the paradigm shift from sampling-based study to population-based real-world study and the examination of interplays between factors at various socioecological levels.

Check out our new article titled “Black businesses matter: A longitudinal study of black-owned restaurants in the COVID-19 pandemic using geospatial big data
” published in Annals of the American Association of Geographers .

GIBD members receive $34,993 in funding from South Carolina Sea Grant Consortium to develop a GIS-based siting tool for South Carolina mariculture site selection. The project team includes Dr. Zhenlong Li (PI), Dr. Cuizhen Wang (Co-PI), and Huan Ning.

The shellfish industry in coastal South Carolina (SC) weighs heavily in culturing and harvesting eastern oyster (Crassostrea virginica) and hard clam (Mercenaria spp.) in tidal creeks and estuaries. Oyster farming in SC is relatively new and is in a smaller scale than those well-established mariculture industries in other Atlantic and Gulf coastal states. However, its economic and public benefits deserve to be recognized. The total oysters farmed in SC has boomed from 139,178 in 2014 to over 1.3 million in 2020 [1]. As filter feeders, these oysters could produce 10 billion gallons of clean water as estimated by SC Department of Natural Resources (SCDNR).

To address the critical need of shellfish siting solutions in South Carolina, this project aims to develop an online GIS-based tool to facilitate the process of selecting an appropriate location for shellfish mariculture lease. The tool will be used by current shellfish growers and potential new growers to successfully locate usable, environmentally and economically beneficial sites as they work through the regulatory process. The longer-term vision of the proposed work is to extend this tool as a sustainable GIS-based mariculture decision-making system that is able to intelligently integrate the multiple siting factors and consider the ever-changing environment with advanced technologies (e.g., spatiotemporal modeling and artificial intelligence) to bolster the healthy and sustainable growth of the South Carolina shellfish aquaculture industry.

The article is open access available at: https://www.tandfonline.com/doi/full/10.1080/19475683.2022.2041725 

Zhang, M., Wang, S., Hu, T., Fu, X., Wang, X., Hu, Y., Halloran B., Li Z. … & Bao, S. (2022). Human mobility and COVID-19 transmission: A systematic review and future directions. Annals of GIS, 1-14.

Abstract: Without a widely distributed vaccine, controlling human mobility has been identified and promoted as the primary strategy to mitigate the transmission of COVID-19. Many studies have reported the relationship between human mobility and COVID-19 transmission by utilizing the spatial-temporal information of mobility data from various sources. To better understand the role of human mobility in the pandemic, we conducted a systematic review of articles that measure the relationship between human mobility and COVID-19 in terms of their data sources, mathematical models, and key findings. Following the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, we selected 47 articles from the Web of Science Core Collection up to September 2020. Restricting human mobility reduced the transmission of COVID-19, although the effectiveness and stringency of policy implementation vary temporally and spatially across different stages of the pandemic. We call for prompt and sustainable measures to control the pandemic. We also recommend researchers 1) to enhance multi-disciplinary collaboration; 2) to adjust the implementation and stringency of mobility-control policies in corresponding to the rapid change of the pandemic; 3) to improve mathematical models used in analysing, simulating, and predicting the transmission of the disease; and 4) to enrich the source of mobility data to ensure data accuracy and suability.

Our new collaborative review paper titled “Human mobility and COVID-19 transmission: a systematic review and future directions” has been accepted for publication by the Annals of GIS.

Abstract:  Without a widely distributed vaccine, controlling human mobility has been identified and promoted as the primary strategy to mitigate the transmission of COVID-19. Many studies have reported the relationship between human mobility and COVID-19 transmission by utilizing the spatial-temporal information of mobility data from various sources. To better understand the role of human mobility in the pandemic, we conducted a systematic review of articles that measure the relationship between human mobility and COVID-19 in terms of their data sources, mathematical models, and key findings. Following the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, we selected 47 articles from Web of Science Core Collection up to September 2020. Restricting human mobility reduced the transmission of COVID-19 spatially, although the effectiveness and stringency of policy implementation vary temporally and spatially across different stages of the pandemic. We call for prompt and sustainable measures to control the pandemic. We also recommend researchers 1) to enhance multi-disciplinary collaboration; 2) to adjust the implementation and stringency of mobility-control policies in corresponding to the rapid change of the pandemic; 3) to improve mathematical models used in analyzing, simulating, and predicting the transmission of the disease; and 4) to enrich the source of mobility data to ensure data accuracy and suability.

Zhang M., Wang S., Hu T., Fu. X., Wang X., Hu Y., Halloran B., Cui Y., Liu H., Bao S., Li Z. (2022). Human mobility and COVID-19 transmission: a systematic review and future directions, Annals of GIS (in press)

Our new article titled “Deep Learning of High-Resolution Aerial Imagery for Coastal Marsh Change Detection: A Comparative Study“ is accepted for publication in the ISPRS International Journal of Geo-Information.

Abstract: Deep learning techniques are increasingly being recognized as effective image classifiers. Aside from their successful performance in past studies, the accuracies have varied in complex environments in comparison with the popularly applied machine learning classifiers. This study seeks to explore the feasibility for using a U-Net deep learning architecture to classify bi-temporal high resolution county scale aerial images to determine the spatial extent and changes of land cover classes that directly or indirectly impact tidal marsh. The image set used in the analysis is a collection of a 1-m resolution collection of National Agriculture Imagery Program (NAIP) tiles from 2009 and 2019 covering Beaufort County, South Carolina. The U-net CNN classification results were compared with two machine learning classifiers, the Random Trees (RT) and the Support Vector Machine (SVM). The results revealed a significant accuracy advantage in using the U-Net classifier (92.4%) as opposed to the SVM (81.6%) and RT (75.7%) classifiers for overall accuracy. From the perspective of a GIS analyst or coastal manager, the U-Net classifier is now an easily accessible nad powerful tool for mapping large areas. Change detection analysis indicated little areal change on marsh extent, though increased land development throughout the county has the potential to negatively impact the health of the marshes. Future work should explore applying the constructed U-Net classifier to coastal environments in large geographic areas, while also implementing other data sources (e.g., LIDAR, multispectral data) to enhance classification accuracy.

Read the full article here: https://www.mdpi.com/2220-9964/11/2/100/pdf

Dr. Zhenlong Li, collaborated with Dr. Shan Qiao from public health, has been awarded a project titled “A novel data-driven approach to empirically link structural racism and health access and utilization in South Carolina” by the UofSC OVPR Racial Justice and Equity Research Program. Other team members include Drs. Xiaoming Li, Bankole Olatosi, and Jiajia Zhang.

Leveraging large place visitation records with high granularity sampled from mobile devices and census data, we propose an integrative big data approach to examine the association between structural racism and disparities in health care access in South Carolina. The proposed research will develop innovative measurement tools to assess the racial/ethnical disparities in health care access and utilization in SC and explore spaciotemporal pattern of such disparities from 2018 to 2021. The findings will provide population level real-world data to address structural racism in SC and improve overall quality of care and population health, especially for African American communities.

Dr. Fengrui Jing received his PhD in GIScience from Sun Yat-sen University in 2021. He also holds a master’s degree in Physical Geography and two undergraduate degrees in Social Work and Psychology. His research focuses on using massive social media data to map neighborhood disorder and fear of crime, and to examine the causal relationship between micro built environment and fear of crime.

Dr. Jing will work with Dr. Zhenlong Li and other team members in GIBD and USC Big Data Health Science Center (BDHSC, https://bigdata.sc.edu) to conduct cutting-edge and innovative interdisciplinary research on geospatial big data analytics (e.g., analyzing massive social sensing data and healthcare records) by using/developing advanced spatiotemporal analysis methods, statistical and predictive models, and computing algorithms and tools in the intersection of Science, data science, and health science.

Welcome Fengrui!

New article titled “Does distance still matter? Moderating effects of distance measures on the relationship between pandemic severity and bilateral tourism demand”, authored by Yang Y., Zhang L., Wu L. and Li Z., has been accepted for publication by Journal of Travel Research (Impact factor: 10.982).

This study aims to investigate the moderating effects of various distance measures on the relationship between relative pandemic severity and bilateral tourism demand. After confirming its validity using actual hotel and air demand measures, we leveraged data from Google Destination Insights to understand daily bilateral tourism demand between 148 origin countries and 109 destination countries. Specifically, we estimated a series of fixed-effects panel data gravity models based on the year-over-year change in daily demand. Results show that a 10% increase in 7-day smoothed COVID-19 cases led to a 0.0658% decline in year-over-year demand change. The moderating distance measures include geographic, cultural, economic, social, and political distance. Results show that long-haul tourism demand was more affected by a destination’s pandemic severity relative to tourists’ place of origin. The moderating effect of national cultural dimensions indulgence versus constraints was also confirmed. Lastly, a discussion and implications for international destination marketing are provided.

Our paper titled “Exploring the spatial disparity of home-dwelling time patterns in the U.S. during the COVID-19 pandemic via Bayesian inference” has been accepted for publication by the Transactions in GIS. 

Abstract: In this study, we aim to reveal hidden patterns and confounders associated with policy implementation and adherence by investigating the home-dwelling stages from a data-driven perspective via Bayesian Inference with weakly informative priors and by examining how home-dwelling stages in the U.S. varied
geographically, using fine-grained, spatial-explicit home-dwelling time records from a multi-scale perspective. At the U.S. national level, two changepoints are identified, with the former corresponding to March 22, 2020 (nine days after the White House declared the National Emergency on March 13) and the latter corresponding to May 17, 2020. Inspections on the U.S. state and county level reveal notable spatial disparity in home-dwelling stages, presumably resulting from the discrepancies in political partisanship, COVID-19 severity, social distancing compliance, re-opening policy, and industry distribution. A pilot study in the Atlanta Metropolitan area at the Census Tract level reveals that the self-quarantine duration and increase in home-dwelling time are strongly correlated with the median household income, echoing existing efforts that document the economic inequity exposed by the U.S. stay-at-home orders. To our best knowledge, our work marks a pioneering effort to explore multi-scale home-dwelling patterns in the U.S. from a pure data-driven perspective and in a statistically robust manner.

Our new paper titled “The promise of excess mobility analysis: measuring episodic-mobility with geotagged social media data” is accepted for publication in the Cartography and Geographic Information Science.

Abstract: Human mobility studies have become increasingly important and diverse in the past decade with the support of social media big data that enables human mobility to be measured in a harmonized and rapid manner. However, what is less explored in the current scholarship is episodic mobility as a special type of human mobility defined as the abnormal mobility triggered by episodic events excess to the normal range of mobility at large. Drawing on a large-scale systematic collection of 1.9 billion geotagged Twitter data from 2017 to 2020, this study contributes the first empirical study of episodic mobility by producing a daily Twitter census of visitors at the U.S. county level and proposing multiple statistical approaches to identify and quantify episodic mobility. It is followed by four case studies of episodic mobility in U.S. national wide to showcase the great potential of Twitter data and our proposed method to detect episodic mobility subject to episodic events that occur both regularly and sporadically. This study provides new insights on episodic mobility in terms of its conceptual and methodological framework and empirical knowledge, which enriches the current mobility research paradigm.

Read the full article here.

Zhenlong Li is co-guest editing a Special Issue entitled “Spatial Analytics for COVID-19 Studies” for the International Journal of Environmental Research and Public Health (ISSN 1660-4601, IF 3.390, http://www.mdpi.com/journal/ijerph).

IJERPH is an open access journal indexed by SCI, SSCI, Scopus, and PubMed. According to Web of Science, IJERPH ranks 118/274 (Q2) in “Environmental Sciences” (SCIE), 68/203 (Q2) in “Public, Environmental, and Occupational Health” (SCIE), and 41/176 (Q1) in “Public, Environmental, and Occupational Health” (SSCI). The median processing time for submissions is less than 45 days, which includes a free English editing service after acceptance of the paper. The article processing charge (APC) is CHF 2300 (Swiss Francs) per accepted paper.

Dear Colleagues,

GIScience for Risk Management in Big Data Era

This Special Issue aims to capture recent efforts and advancements in harnessing the power of GIScience for risk management in the big data era.

The first group of possible topics is to inspire potential authors to deal with basic and new trends related to the big data era. The contribution of novel approaches to spatial data collection (social networks, sensors, citizen science, VGI, etc.), disaster big data processing and sharing, real-time data-centric intelligence based on sensors, harmonization of heterogeneous data into a single structure, cybersecurity of geographical information systems and others, is welcomed, along with analyses and commentary.

The second thematic block will cover cartography and GIS theories such as mobile disaster cartography, concepts, ontologization and standardization, cross-cultural aspects of disaster cartography, investigation of the psychological condition of end-users given by their personal character and situation, and the psychological condition of rescued persons are offered together with questions that are still open on the mapping methodologies and technologies for EW&CM from children and senior perspectives.

The third group of topics aims to address mapping and visualization techniques. Dynamic and real-time cartographic visualization concepts and techniques for enhanced operational activities for selected EW, DRM, and DRR purposes are highlighted. Included in the same group are both virtual environments for EW, DRM, and DRR as well as 3D analysis and visualization of disaster events.

The last group of topics is devoted to services and applications, and may include analyses and descriptions of location-based services for emergencies (web services, etc.), multimodal emergency positioning, mapping based on social big data, internet of things for solutions and visualizations, and disaster chain modeling.

In particular, potential inspiring topics for authors include the following:

Big data
Novel approaches to spatial data collection (social networks, sensors, citizen science, VGI, etc.)
Geospatial big data computing, analytics, and sharing for disaster management
Real-time data-centric intelligence based on sensors for purposes of DRM and DRR harmonization and homogenization of heterogenous data.
Searching and calculations of anomalies in geospatial big data in DRM and DRR process
Cartographic use of remotely sensed and other geospatial data for early warning, DRM, and DRR
Cybersecurity of geographical information systems (of data flows from sensor networks to GIS platforms)
Cartography and GIS theories

Mobile disaster cartography
Concepts, ontologization, and standardization for early warning, hazard, risk, and vulnerability mapping
Mechanisms of command and control systems integration
Cross-cultural aspects of disaster cartography (traditions, universality, and conventions and their integration)
Investigation of the psychological condition of end-users given by their personal character and situation and the psychological condition of rescued persons
Mapping methodologies and technologies for EW&CM from the perspectives of children and seniors. Designing, understanding, and using maps for EW, DRM, and DRR for children and seniors

Mapping and visualization techniques
Dynamic and real-time cartographic visualization concepts and techniques for enhanced operational early warning and DRM activities for selected purposes (various government levels, inter-state cooperation, first aid, etc.)
Virtual environments for EW and DRR (geographic, indoors, underground, etc.)
3D disaster (floods, fires, slides, tsunamis, etc.) analysis and visualization

Services and application
Location-based service for emergencies
Multimodal emergency positioning
Disaster risk analyses and mapping using social big data
Internet of things (IoT) in disaster solutions and visualizations
Disaster chain modeling

Special Issue Editors

A new article led by Huan Ning, titled “Exploring the Vertical dimension of Street View Image Based on Deep Learning: A Case Study on Large-scale Building Flooding Risk Assessment”, has been accepted for publication in the International Journal of Geographical Information Science, a flagship journal in GIScience.

Congratulations, Huan!

Review article entitled “Human Mobility Data in the COVID-19 Pandemic: Characteristics, Applications, and Challenges” accepted for publication by the International Journal of Digital Earth (2020 Impact Factor: 3.538)

Read preprint here.

Zhenlong Li gave an invited presentation titled “Measuring Human Mobility Dynamics and Place Connectivity Using Big Social Media Data” at Geospatial Science and Human Security Division Director Seminar Series of the Oak Ridge National Laboratory on June 24, 2021.

Abstract: Understanding human mobility dynamics among places provides fundamental knowledge regarding their interactive gravity, benefiting a wide range of applications in need of knowledge in human spatial interactions. The ongoing COVID-19 pandemic uniquely highlights the need for monitoring, measuring, and predicting human movement at various geographic scales from local to global. This talk first introduces our recent effort in quantifying global human movement using billions of geotagged tweets coupled with big data computing, and then presents a global multi-scale place connectivity index (PCI) derived from such movement. Two application examples are followed to exemplify the utility of PCI as a factor in 1) predicting the spatial spread of COVID-19 during the early stage, and 2) predicting hurricane evacuation destination choices.

A new article titled “Introducing Twitter Daily Estimates of Residents and Non-Residents at the County Level”, authored by Yago Martin, Zhenlong Li, Yue Ge, and Xiao Huang, is published in Social Sciences.

Abstract: The study of migrations and mobility has historically been severely limited by the absence of reliable data or the temporal sparsity of available data. Using geospatial digital trace data, the study of population movements can be much more precisely and dynamically measured. Our research seeks to develop a near real-time (one-day lag) Twitter census that gives a more temporally granular picture of local and non-local population at the county level. Internal validation reveals over 80% accuracy when compared with users’ self-reported home location. External validation results suggest these stocks correlate with available statistics of residents/non-residents at the county level and can accurately reflect regular (seasonal tourism) and non-regular events such as the Great American Solar Eclipse of 2017. The findings demonstrate that Twitter holds the potential to introduce the dynamic component often lacking in population estimates. This study could potentially benefit various fields such as demography, tourism, emergency manage

ment, and public health and create new opportunities for large-scale mobility analyses.

Read full article here: https://www.mdpi.com/2076-0760/10/6/227/pdf