VietMed-NER

该数据集名为VietMed-NER，是全球最大的口语命名实体识别（NER）数据集，专注于现实世界中的医疗对话，涵盖了18种不同的实体类型。该数据集采用人工标注的黄金标准实体标签，并将这些标签与相应的对话语句关联起来，构建了一个知识图谱。它是目前规模最大的NER数据集，包含了多种实体类型，其任务是对命名实体进行识别。

This dataset, named VietMed-NER, is the world's largest spoken named entity recognition (NER) dataset focused on real-world medical dialogues, encompassing 18 distinct entity types. It adopts manually annotated gold-standard entity labels, links these labels with their corresponding dialogue utterances to construct a knowledge graph. As the largest NER dataset to date with various entity types, its task is to perform named entity recognition.

搜集汇总

背景与挑战

背景概述

VietMed-NER是全球最大的口语命名实体识别数据集，专注于医疗对话，涵盖18种实体类型，采用人工标注构建了知识图谱。该数据集以现实世界医疗对话为基础，支持大规模NER任务，为医疗领域自然语言处理提供了高质量资源。