LEAP corpus
收藏arXiv2023-10-31 更新2024-08-06 收录
下载链接：
http://arxiv.org/abs/2310.20260v1
下载链接
链接失效反馈官方服务：
资源简介：
LEAP corpus是由曼彻斯特大学计算机科学系开发的异构数据集，包含从一本国际象棋教科书中收集的1164个句子，讨论了91场比赛中的策略移动。数据集不仅包括结构化的国际象棋移动和棋盘状态，还包括非结构化的文本描述。该数据集旨在帮助自然语言处理国际象棋代理评估来自异构国际象棋知识源的移动。通过标注相关性和情感标签，数据集支持基于情感分析的国际象棋移动评估，并已用于训练和评估多种基于Transformer的模型。
The LEAP corpus is a heterogeneous dataset developed by the Department of Computer Science, University of Manchester. It contains 1,164 sentences collected from a chess textbook, which discuss strategic moves across 91 chess games. This dataset not only includes structured chess moves and board states, but also unstructured textual descriptions. It is designed to help natural language processing (NLP) chess agents evaluate moves derived from heterogeneous chess knowledge sources. Through the annotation of relevance and sentiment labels, the dataset supports sentiment analysis-based chess move evaluation, and has been used to train and evaluate multiple Transformer-based models.
提供机构：
曼彻斯特大学计算机科学系
创建时间：
2023-10-31
