Title: Prediction and Analysis of Urban Mobility Based on Attention Mechanism and Geographic Information Embedding

Title: Exploring the nonlinear effects of greenery on active travel among the ageing population

This project aims to develop a geographical model-based intelligent prediction system for crowd flow simulation and emergency decision support in disasters, effectively leveraging Digital Twin data. To accurately predict crowd flow changes, we will construct a sophisticated prediction model that integrates comprehensive data from geographical environment, disaster dynamics, and crowd movement trajectories. Digital Twin technology will be strategically utilized as a rich data source and a powerful virtual simulation platform.

Title: LandGPT: a multimodal large language model for parcel-level land use classification with multi-source data

Title: A causal intervention framework for synthesizing mobility data and evaluating predictive neural networks