LA-MGFM: A legal judgment prediction method via sememe-enhanced graph neural networks and multi-graph fusion mechanism

Qihui Zhao, Tianhan Gao, Nan Guo

Abstract: Highlights•Five kinds of text graphs are constructed for the legal judgment prediction task.•A GGNN is proposed to incorporate the law articles and Sememe information.•A multi-graph fusion mechanism is proposed to merge the feature of different graphs.•The confusion problem of charges is effectively alleviated.•The model achieves competitive results, including in the few-shot condition.