This repository contains the promotional materials for the GeoTopo-Net model presented in our paper, Predicting short-term urban bike sharing demand in a coupled continuous and network space.

To predict local demand, is proximity enough, or do the pathways people travel also matter? Our AI model (GeoTopo-Net) proves that including the layout of urban networks is key. We find that metro lines give a greater accuracy boost than cycling paths, especially in urban centers and near stations.

Fig. 1. (A) Overall design of GeoTopo-Net. In the implementation of the model, two variants of the GeoTopo block are designed: (B) Sequential structure; (C) Parallel structure.

If you consider it useful for your research or development, please consider citing our paper.

@article{TBS2026GeoTopoNet,
    author = {Shen Liang and Yang Xu and Guangyue Li and Xiaohu Zhang and Qiuping Li},
    title = {Predicting short-term urban bike sharing demand in a coupled continuous and network space},
    journal = {Travel Behaviour and Society},
    volume = {42},
    pages = {101152},
    year = {2026},
    publisher = {Elsevier},
    doi = {10.1016/j.tbs.2025.101152},
}