On region-level travel demand forecasting using multi-task adaptive graph attention network

Jian Liang, Jinjun Tang, Fan Gao, Zhe Wang, Helai Huang

Abstract: Highlights•We propose a multi-task adaptive recurrent graph attention network to predict travel demand.•We use the BM25 algorithm to characterize region function.•We design a prior knowledge-driven graph learning mechanism.•The proposed method outperforms the state-of-the-art approaches in prediction accuracy.