JoelMba/DEFT2021_CAS_annotations_combined_v2

该数据集是一个用于命名实体识别（NER）任务的数据集，特别关注于疾病（Disorders）相关的实体识别。数据集包含训练集、验证集和测试集，分别有2279、780和1215个示例。每个示例包含id、tokens、ner_tags（标注的疾病实体标签）、ner_tag_labels和document_id等特征。ner_tags使用BIO标注方案，其中B-Disorders表示疾病实体的开始，I-Disorders表示疾病实体的内部，O表示非实体部分。

This dataset is tailored for named entity recognition (NER) tasks, with a particular emphasis on the identification of disease-related entities. The dataset is split into training, validation, and test sets, which contain 2279, 780, and 1215 instances respectively. Each instance includes features such as id, tokens, ner_tags (annotated disease entity labels), ner_tag_labels, and document_id. The ner_tags adopt the BIO tagging scheme, where B-Disorders represents the start of a disease entity, I-Disorders refers to the internal part of a disease entity, and O denotes non-entity segments.