LegalCon
收藏资源简介:
LegalCon是一个包含1,063个标记法庭对话的数据集,用于检测操纵、识别主要操纵者以及分类操纵技术。数据集由Manipal University Jaipur创建,旨在解决法庭对话中操纵的检测和分析问题。数据集来源于美国各级法院的庭审记录,包括最高法院、家庭支持法院、审判法院和小额申诉法院。该数据集通过人工标注,确保了数据的准确性和可靠性,并为法庭语言分析提供了宝贵资源。
LegalCon is a dataset consisting of 1,063 annotated court dialogues, developed for three core tasks: detecting manipulation, identifying primary manipulators, and classifying manipulation techniques. It was created by Manipal University Jaipur to address the detection and analysis of manipulation in court dialogues. The dataset is sourced from trial transcripts of U.S. courts at all levels, including the Supreme Court, Family Support Court, Trial Court, and Small Claims Court. Manual annotation is employed to ensure the accuracy and reliability of the data, making it a valuable resource for forensic linguistic analysis.




