RAAMove
收藏资源简介:
RAAMove是由北京理工大学外国语学院等机构开发的一个大型多领域语料库,专注于研究文章摘要中的移动结构注释。该数据集包含来自人工智能、通信工程和机械工程三个科学领域的33,988个注释实例。数据集的构建分为两个阶段:首先由专家注释者手动注释高质量数据,随后使用基于BERT的模型进行自动注释,并由专家进行修正。RAAMove旨在促进移动分析和自动移动识别,可作为移动分析、英语语言教学和写作以及自然语言处理中与移动/话语相关任务的重要资源。
RAAMove is a large-scale multi-domain corpus developed by the School of Foreign Studies of Beijing Institute of Technology and other institutions, focusing on the annotation research of move structures in academic abstracts. This corpus contains 33,988 annotated instances from three scientific disciplines: artificial intelligence, communication engineering, and mechanical engineering. The construction of the dataset is divided into two stages: first, high-quality data is manually annotated by expert annotators, then automatically annotated using a BERT-based model, and finally revised by experts. RAAMove aims to facilitate move analysis and automatic move recognition, and can serve as an important resource for move analysis, English language teaching and writing, and move/discourse-related tasks in natural language processing.




