comma-project/comma-jsonl
收藏Hugging Face2025-11-21 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/comma-project/comma-jsonl
下载链接
链接失效反馈官方服务：
资源简介：
CoMMA是一个包含超过2.5亿个词汇的大型中世纪手稿语料库，这些词汇来自超过23000份拉丁语和古法语手稿，覆盖了801至1600年的时间范围。该语料库提供了原始的非标准化文本，适用于自然语言处理模型的训练与评估，以及历史语言学和数字人文等领域的研究。
CoMMA is a large-scale corpus of digitized medieval manuscripts containing over 250 million tokens from more than 23,000 manuscripts in Latin and Old French, covering the period from 801 to 1600 CE. The corpus provides raw, non-normalized text suitable for training and evaluation of NLP models, as well as research in historical linguistics and digital humanities.
提供机构：
comma-project
