ECHO (Earth-Climate-Health Outlook) Lecture Series #9

Abstract: In the era of urban big data, cities are increasingly understood as complex and interconnected networks shaped by spatial relationships, flows, and human interactions. This presentation explores how graph modelling and spatially explicit Geospatial Artificial Intelligence (GeoAI) can advance urban analytics by moving beyond conventional map-based representations towards networked understandings of urban systems. Emphasising that “spatial is special”, the talk demonstrates how geographic priors — including contiguity, proximity, hierarchy, and spatial heterogeneity — can be directly embedded within Graph Neural Networks (GNNs). By representing urban environments as graph structures that reflect cities' underlying spatial logic, these approaches enable more interpretable, scalable, and policy-relevant urban analysis. Drawing on recent applications in GeoAI and urban analytics, the presentation illustrates how geographically informed graph learning can generate deeper insights into complex urban challenges, including socio-spatial inequality, environmental exposure, mobility patterns, and infrastructure resilience.

About the Speaker:

Dr Pengyuan Liu is currently working as a Lecturer (Assistant Prof) in Digital Planning at the University of Glasgow. His expertise lies in the field of quantitative urban geography, with a specialisation in GeoAI and urban digital twin research. He holds a PhD in Geography from the University of Leicester, UK. Before joining the University of Glasgow, he served as a Senior Postdoctoral Researcher and Module Coordinator at the Future Cities Laboratory, Singapore-ETH Centre. In 2023, he contributed as a Lecturer in Human Geography and Urban Planning at Nanjing University of Information Science and Technology. His academic journey also includes valuable postdoctoral research experiences at the Urban Analytics Lab, National University of Singapore and the University of Helsinki.