D2GCLF: Document-to-Graph Classifier for Legal Document Classification
Qiqi Wang, Kaiqi Zhao, Robert Amor, Benjamin Liu, Ruofan Wang
Abstract
Legal document classification is an essential task in law intelligence to automate the labor-intensive law case filing process. Unlike traditional document classification problems, legal documents should be classified by reasons and facts instead of topics. We propose a Document-to-Graph Classifier (D2GCLF), which extracts facts as relations between key participants in the law case and represents a legal document with four relation graphs. Each graph is responsible for capturing different relations between the litigation participants. We further develop a graph attention network on top of the four relation graphs to classify the legal documents. Experiments on a real-world legal document dataset show that D2GCLF outperforms the state-of-the-art methods in terms of accuracy.
BibTeX
@inproceedings{wang-etal-2022-d2gclf,
title = "{D}2{GCLF}: Document-to-Graph Classifier for Legal Document Classification",
author = "Wang, Qiqi and
Zhao, Kaiqi and
Amor, Robert and
Liu, Benjamin and
Wang, Ruofan",
editor = "Carpuat, Marine and
de Marneffe, Marie-Catherine and
Meza Ruiz, Ivan Vladimir",
booktitle = "Findings of the Association for Computational Linguistics: NAACL 2022",
month = jul,
year = "2022",
address = "Seattle, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-naacl.170/",
doi = "10.18653/v1/2022.findings-naacl.170",
pages = "2208--2221"
}