LEDGAR

LEDGAR是一个包含合同法律条款的多标签语料库，它是从公共领域（美国证券交易委员会提交的文件）中抓取和整理的，据我们所知，这是同类中首个免费可用的语料库。由于该语料库是半自动构建的，我们采用了并讨论了多种去噪方法。由于该语料库包含超过60,000份合同中的近100,000个条款，并且标注了超过12,000个标签，我们认为它对于法律自然语言处理（Legal NLP）、大规模或极端文本分类以及法律研究等领域的研究都具有重要意义。LEDGAR is a multi-label corpus containing contractual legal clauses, which was collected and curated from public-domain documents filed with the U.S. Securities and Exchange Commission (SEC). To the best of our knowledge, it is the first freely available corpus of its kind. As the corpus was constructed semi-automatically, we adopted and discussed multiple denoising methods. Comprising nearly 100,000 clauses from over 60,000 contracts and featuring over 12,000 annotated labels, we deem it highly significant for research in fields such as Legal Natural Language Processing (Legal NLP), large-scale or extreme text classification, and legal research.

搜集汇总

背景与挑战

背景概述

LEDGAR是一个基于美国证券交易委员会合同文件构建的大规模多标签语料库，专注于法律条款分类。它包含超过60,000份合同中的近100,000个条款，并标注了超过12,000个标签，是首个免费可用的同类资源。该数据集采用半自动构建和去噪处理，对法律自然语言处理和大规模文本分类研究具有重要意义。