Postgraduate research students

Fengjiao Li

E-mail:f.li.3@research.gla.ac.uk 

 

Personal Website 

ORCID iDhttps://orcid.org/0009-0006-9075-0818

Research title: Towards GeoAI-Enhanced Mobility-Based Health Risk Analysis: Embedding Spatial Intelligence into Graph Neural Networks for Dynamic Population Modelling

Research summary

My research lies at the intersection of GeoAI, spatial data science, and machine learning. I am interested in developing new methods for spatial-temporal graph learning, with a particular focus on representation learning, adaptive graph learning, and interpretable GeoAI for modelling complex spatial systems.

My current research investigates how different forms of geographic information, including spatial proximity, mobility patterns, transportation networks, and socioeconomic characteristics, can be effectively represented, adaptively learned, and interpreted within graph learning frameworks. By integrating these heterogeneous spatial relationships, I aim to improve the accuracy, robustness, and interpretability of spatial prediction models while providing deeper insights into the underlying spatial processes.

I evaluate these methods across public health and transportation applications, with the broader goal of developing transferable GeoAI approaches for urban forecasting and spatial decision support.

 

Publications

List by: Type | Date

Supervisors

Grants

  • 2024-2028, University of Glasgow - The China Scholarship Council (CSC) Co-operative Scholarship.

  • 2026, Centre for Data Science Conference and Training fund，£1345

  • 2026, International Conference on Geospatial Artificial Intelligence, Travel grants, €300

Conferences

Li, F., Wu, M., & Basiri, A. (2025, November 20). Mobility vs. Contiguity: Spatially Explicit Graph Neural Networks for COVID-19 Forecasting. The 6th Spatial Data Science Symposium (SDSS 2025). https://doi.org/10.5281/zenodo.17660772

Li, F., Wu, M., & Basiri, A. (2026, April 8). Spatially Explicit Graph Neural Networks for Epidemic Forecasting: Evidence from Socioeconomic Embeddings. The 1st International Conference on Geospatial Artificial Intelligence (GeoAI 2026), Ghent, Belgium. https://doi.org/10.5281/zenodo.19475486

Han, T., Li, F., Chen, . chunsong ., Huang, H., Wu, M., & Chen, Y. (2026, May 20). Spatially-Weighted CLIP for Street-View Geo-localization. The 1st International Conference on Geospatial Artificial Intelligence (GeoAI 2026), Ghent, Belgium. https://doi.org/10.5281/zenodo.20313536

Teaching

Teaching Assistant：

School of Mathematics and Statistics

  • Maths 2T, 2025
  • Maths 2D, 2025

School of Geographical & Earth Sciences

  • Geography 2 STATS-GIS, 2026
  • GEOG5015: Web and Mobile Mapping, 2026

Research Assistant：

RA, in Social Sciences Administration within the College of Social Sciences. 2025.

RA, in  School of Geographical & Earth Sciences, 2026 

Additional information

Education：

  • M.Sc. in Applied Statistics (Biostatistics), University of Liverpool, 2024
  • B.Sc. in Applied Mathematics and Statistics, University of Wisconsin, 2022
  • B.Sc. in Economic Statistics,Suzhou University of Technology, 2022

Professional Experience：

  • Biostatistician

    Zenith CRO, Shanghai, China
    12/2023 – 09/2024

    Transcenta, Suzhou, China
    06/2023 – 10/2023

  • SAS Analyst

        ClinChoice，Nanjing，China

Skills: Python, R, SAS, PyTorch