Spatial-temporal knowledge graph network for event prediction

Zepeng Huai, Dawei Zhang, Guohua Yang, Jianhua Tao

Abstract: Highlights•We construct a novel spatial-temporal knowledge graph for event prediction, which both holds the trans-regional influence and the time sequence pattern.•We propose a continuous-time dynamic graph neural network to simulate and forecast the evolving process of entities.•We conduct comprehensive experiments on two public datasets, which proves the effectiveness and interpretability of our proposed model.