MVE-FLK: A multi-task legal judgment prediction via multi-view encoder fusing legal keywords

作者：

Highlights：

• Taking full advantage of existing law articles is critical to the task of legal judgment prediction.

• A novel multi-task legal judgment prediction framework via multi-view encoder fusing legal (MVE-FLK) is proposed for legal judgment prediction (LJP).

• MVE-FLK is appropriate for the task of multi-class classification with multi-label learning.

• The experimental results verify the effectiveness of MVE-FLK for the task of LJP.

摘要

•Taking full advantage of existing law articles is critical to the task of legal judgment prediction.•A novel multi-task legal judgment prediction framework via multi-view encoder fusing legal (MVE-FLK) is proposed for legal judgment prediction (LJP).•MVE-FLK is appropriate for the task of multi-class classification with multi-label learning.•The experimental results verify the effectiveness of MVE-FLK for the task of LJP.

论文关键词：Legal judgment prediction,Multi-class classification,Multi-label learning,Multi-task learning,Word and sentence encoder

论文评审过程：Received 23 June 2021, Revised 6 December 2021, Accepted 11 December 2021, Available online 17 December 2021, Version of Record 10 January 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107960