Rui Deng

rui.deng@glasgow.ac.uk

Research title: Advancing Transformer-based models for understanding geospatial tabular data in urban systems

Research Summary

Currently, I'm interested in developing efficient and effective deep learning models for geospatial tabular data, with a particular focus on transformer-based context modeling methods.

I apply various GeoAI techniques and explainable AI (XAI) methods to analyze the factors that influence human perception of urban environments and to support the design of sustainable urban systems.

Publications

Selected Publications:

Deng, R., Guan, Y., Cai, D., Yang, T., Fraedrich, K., Zhang, C., Tang, J., Liao, Z., Wei, Z. and Guo, S., 2023. Supervised versus semi-supervised urban functional area prediction: uncertainty, robustness and sensitivity. Remote Sensing, 15(2), p.341.

Deng, R., Li, Z. and Wang, M., 2025. GeoAggregator: An Efficient Transformer Model for Geo-Spatial Tabular Data. arXiv preprint arXiv:2502.15032 (Accepted in the 39th AAAI Conference on Artificial Intelligence).

Supervisors