Why places feel the way they do
A two-stage GeoAI framework that parses free-form explanations into structured perceptual units to quantify how specific elements shape safety, beauty, and liveliness across cities.
PhD student, NYU Wagner & NYU Shanghai. I build human-centered urban digital twins that integrate multi-modal data, large language models, and spatial reasoning to explain and improve how people experience cities.
Explainable Perception
Session 13.29: Technology, Society and Analytical Methods
Thu, Oct 23, 2025
11:00 AM – 12:30 PM • Orchestra D
Street Connectivity
Session 15.14: Urban Design
Thu, Oct 23, 2025
3:30 PM – 5:00 PM • Marquette 7
Hi, I am Tim Guangyu Wu, a PhD student at NYU Wagner and NYU Shanghai, and a researcher with the NYU Shanghai Urban Lab. My advisors are Professor ChengHe Guan and Professor Zhan Guo. I received my Master’s degree in Data Science from the Columbia Data Science Institute and my Bachelor’s degree in Data Science and Finance from NYU Shanghai. Previously, I worked at Columbia University’s Center for Spatial Research, where I led the data science work for the Mapping Historical New York project.
A two-stage GeoAI framework that parses free-form explanations into structured perceptual units to quantify how specific elements shape safety, beauty, and liveliness across cities.
National-scale analysis separating regulatory impacts on public and private subdivisions with improved node classification and temporal trends for policy insight. This site is currently migrating to a large language model agent-based interface—more project details will be available soon. Stay tuned.
Email: gw923@nyu.edu
GitHub: @timwgy
I’m always open to chats and collaborations—feel free to reach out or connect at ACSP in Minneapolis!