Neural Legal Judgment Prediction in English

• DOI: 10.18653/v1/p19-1424
• Publication Date: 2019-09-12T18:39:56Z
• Authors: Ilias Chalkidis; Ion Androutsopoulos; Nikolaos Aletras
• Keywords: Feature (linguistics); Binary classification; Predictive modelling

Legal judgment prediction is the task of automatically predicting outcome a court case, given text describing case’s facts. Previous work on using neural models for this has focused Chinese; only feature-based (e.g., bags words and topics) have been considered in English. We release new English legal dataset, containing cases from European Court Human Rights. evaluate broad variety establishing strong baselines that surpass previous three tasks: (1) binary violation classification; (2) multi-label (3) case importance prediction. also explore if are biased towards demographic information via data anonymization. As side-product, we propose hierarchical version BERT, which bypasses BERT’s length limitation.