tner/conll2003
收藏资源简介：
CoNLL-2003 NER数据集是一个用于命名实体识别任务的数据集，主要来源于新闻领域。该数据集包含四种实体类型：ORG（组织）、PER（人名）、LOC（地点）和MISC（其他）。数据集的结构包括训练集、验证集和测试集，分别包含14041、3250和3453个样本。数据集的版权归Reuters Ltd和/或Thomson Reuters所有，使用时需要签署相关协议。
The CoNLL-2003 NER dataset is a dataset for named entity recognition (NER) tasks, primarily sourced from the news domain. It includes four entity types: ORG (Organization), PER (Person), LOC (Location), and MISC (Miscellaneous). The dataset is structured into training, validation, and test sets, which contain 14,041, 3,250, and 3,453 samples respectively. The copyright of this dataset belongs to Reuters Ltd and/or Thomson Reuters, and relevant agreements must be signed prior to its usage.
数据集概述
数据集基本信息
- 名称: CoNLL-2003
- 语言: 英语
- 许可证: 其他
- 多语言性: 单语
- 大小: 10K<n<100K
- 任务类别: 词元分类
- 任务ID: 命名实体识别
- 美观名称: CoNLL-2003
数据集详细描述
- 领域: 新闻
- 实体数量: 3
- 实体类型:
ORG,PER,LOC,MISC
数据集结构
数据实例
json { tags: [SOCCER,-, JAPAN, GET, LUCKY, WIN, ,, CHINA, IN, SURPRISE, DEFEAT, .], tokens: [0, 0, 5, 0, 0, 0, 0, 3, 0, 0, 0, 0] }
标签ID
python { "O": 0, "B-ORG": 1, "B-MISC": 2, "B-PER": 3, "I-PER": 4, "B-LOC": 5, "I-ORG": 6, "I-MISC": 7, "I-LOC": 8 }
数据分割
| 名称 | 训练 | 验证 | 测试 |
|---|---|---|---|
| conll2003 | 14041 | 3250 | 3453 |
许可证信息
数据集的版权信息来源于CoNLL2003 shared task页面，具体版权协议需通过Organizational agreement和Individual agreement了解。
引用信息
bibtex @inproceedings{tjong-kim-sang-de-meulder-2003-introduction, title = "Introduction to the {C}o{NLL}-2003 Shared Task: Language-Independent Named Entity Recognition", author = "Tjong Kim Sang, Erik F. and De Meulder, Fien", booktitle = "Proceedings of the Seventh Conference on Natural Language Learning at {HLT}-{NAACL} 2003", year = "2003", url = "https://www.aclweb.org/anthology/W03-0419", pages = "142--147", }
