Chinese Essay Argument Mining Corpus (CEAMC)
收藏资源简介:
CEAMC数据集是由华东师范大学的研究团队创建的,包含226篇中国高中生议论文,每篇论文都被标注了4种粗粒度和10种细粒度的句子级论证成分。数据集还包括了作文的得分信息。研究团队对论证成分之间的关系进行了详细的标注,从垂直和水平两个维度分析了论证结构,共标注了4837个关系。这些标注数据为论证分析、写作质量评估和文本生成等下游NLP任务提供了重要的支持。
The CEAMC dataset was created by a research team from East China Normal University. It contains 226 argumentative essays written by Chinese high school students, each annotated with 4 coarse-grained and 10 fine-grained sentence-level argumentative components. The dataset also includes scoring information for each composition. The research team conducted detailed annotation of the relationships between argumentative components, analyzed the argumentation structure from both vertical and horizontal dimensions, and annotated a total of 4,837 relationships. These annotated data provide important support for downstream NLP tasks such as argument analysis, writing quality assessment, and text generation.

- 1Towards Comprehensive Argument Analysis in Education: Dataset, Tasks, and Method华东师范大学 · 2025年



