Loading...

Geo-Social Lab

Connecting Nodes to Knowledge



Geo-Social Lab is home to research projects aimed at developing innovative computational and visual tools to analyze massive and complex geo-social networks and geospatial processes that drive our economy, society and the environment.

Our aim is to help address research questions and real-world problems in diverse areas such as human mobility and migration, hazards and disasters, city planning and transportation, spatial and historical demography, global trade, communication and information networks, public discourse and information diffusion, etc.

Team

Dr. Caglar Koylu

Assistant Professor

Geographical and Sustainability Sciences

University of Iowa

Caglar Koylu

Director

Hoeyun Kwon

Ph.D. in Geography / GIScience

Hoeyun Kwon

Ph.D. Student

Beichen Tian

Ph.D. in Geography / GIScience

Beichen Tian

Ph.D. Student

Geng Tian

M.A. in Geography / GIScience

Geng Tian

Alumni

Angelina Evans

B.S. in Geography / Minor in Computer Science

Angelina Evans

B.S. Student

Kaitlyn Hom

High School Student

Kaitlyn Hom

High School Student

Mark Rifkin

High School Student

Mark Rifkin

High School Student

Research


FlowMapper: A web-based tool for designing origin-destination flow maps
Flow density estimation for multi-scale flow mapping and generalization
Flow density estimation for multi-scale flow mapping and generalization
Understanding how people perceive and use flow maps
Network measures to calculate vulnerability in spatial networks
Smoothing spatial networks to identify locational characteristics
Geo-social network modeling for understanding interpersonal online communication patterns
Reciprocal mention network on Twitter
Comparing population pyramids: Rootsweb family trees vs 1880 Census
Representativeness of Rootsweb family trees for population demographics
Measuring and visualizing connectedness from family trees
Extracting family migration from population-scale family trees
CarSenToGram: Geovisual analytics for understanding the evolution of public discourse and sentiment
CarSenToGram: Geovisual analytics for understanding the evolution of public discourse and sentiment
Geovisual analytics to track evolution of topics on Twitter
Flow visual analytics for evaluating the efficiency of health care systems
Geovisual analytics for environmental decision-making
Deep learning for disaster management
Deep learning for disaster management
Big social media data analytics for disaster management
Image object detection and kernel density estimation for detecting human activity patterns
Image object detection and kernel density estimation for detecting human activity patterns


Spatial interactions are the movements of tangible and intangible phenomena such as people, goods, vehicles and information between locations. Spatial interactions form location-to-location networks. Analysis of spatial interaction networks is critical for diverse domains such as migration, epidemics, health care, transportation, and economy. Flow maps are commonly used to visualize spatial interactions and facilitate the understanding of patterns of spatial flows and the corresponding spatial context.

Understand the spatial, temporal and relational (network) aspects of dynamic and geographically-embedded social networks. A dynamic geo-social network evolves (changes) over space and time as the actors of the network move (migrate), new actors are added or removed, and relationships between the actors develop and change over time.

Support a human-centered process for pattern searching and knowledge construction through geovisual analytics and usability evaluation.

Provide a data-driven understanding of the complex system of human society and the physical environment through geographic knowledge discovery enabled by the state of the art machine learning and deep learning methods and their applications in spatial data science

Big data analytics for social media and networking applications, e.g., geospatial semantics, natural language processing, topic modeling, sentiment analysis, machine learning and deep learning methods.

Applications and Funding

We offer a diverse set of degrees and research experiences for graduate, undergraduate and high school students through the Department of Geographical and Sustainability Sciences, and the Interdisciplinary Graduate Program at Informatics (IGPI), and the Secondary Student Training Program (SSTP) at the University of Iowa.

  1. Geography graduate degrees are classified as STEM with the CIP code 45.0702 (Cartography and Geographic Information Systems).
  2. For students from Computer Science and other related disciplinary background, Interdisciplinary Graduate Program in Informatics (IGPI)is a great fit.
  3. For undergraduate students we offer a 5-year U2G Programwith a B.S. Geography/GIScience and M.S. degree in Informatics.
  4. For high school students we offera multi-week residential summer research program.

Graduate teaching and research fellowships, and assistantships are available for competitive students. Before applying, please contact Caglar Koyluwith your brief research interests and CV attached. Competitive students will be invited for a Skype interview. The interview starts with a 6 minute, 40 second, a Pecha Kucha style presentation. Your presentation should focus on your current work, future research goals and interests, and how those intersect with the Lab's research agenda.


We are always looking for MS / PhD or PhD students with a variety of research interests such as:


  • Spatial interaction analysis and visualization (e.g., network measures, spatial community detection, spatial interaction modeling and diffusion models, flow mapping and clustering)
    • Spatial data mining (e.g., clustering, association rule mining, machine learning and deep learning)
  • Geovisual analytics and human computer interaction (e.g., interactive cartography, flow mapping, coordinated views, and utility and usability evaluation)
  • Big data analytics for social media and networking applications (e.g., geospatial semantics, natural language processing, topic modeling and sentiment analysis)
  • Application areas such as human migration and mobility, geographically-embedded social networks (e.g., interpersonal communication, family tree), patient mobility, human communication, transportation, and energy networks; movement and pass networks in sports (e.g., soccer, basketball); analysis of social media data; social sensing and networks for disaster response, recovery and resilience

In addition to the above research interests, students should have, or be interested in developing, ability in:

  • Geospatial programming in Java and/or Python
  • Web-based GIS and Geovisualization: JavaScript, D3, React, etc.
  • Statistical computing in R
  • Interactive computing using Jupyter andObservableNotebooks
  • GIS Software such as ArcGIS or QGIS
  • Database management systems such as PostgreSQL/PostGIS
  • High performance computing, Spark, Hadoop and big data storage and management systems (MongoDB)

Check outFlowMapper