c-ho/2026-05-28_ub_opus_bll_ner_annotation
收藏资源简介:
该数据集是一个用于命名实体识别(NER)任务的数据集,包含文本行及其分词结果,以及标注的实体信息。每个样本包括文档ID(doc_id)、原始文本行(line_text)、分词后的词元序列(tokens),以及NER结果列表(ner_results),其中每个实体标注包含实体类型(entity_type)、字符级和词元级的起止索引、实体类别ID(bll_class_id)和字面表达(bll_class_literal)。数据集仅提供训练分割(train),包含15,665个样本,总大小约为26.4 MB。
This dataset is designed for Named Entity Recognition (NER) tasks, containing text lines along with tokenized sequences and annotated entity information. Each sample includes a document ID (doc_id), the original text line (line_text), a sequence of tokens (tokens), and a list of NER results (ner_results), where each entity annotation comprises entity type (entity_type), character and token-level start and end indices, entity class ID (bll_class_id), and literal expression (bll_class_literal). The dataset only provides a training split (train) with 15,665 examples and a total size of approximately 26.4 MB.



