Deactivated Concepts Medical Term Similarity Dataset
收藏arXiv2025-09-30 收录
下载链接：
https://www.snomed.org/snomed-ct/five-step-briefing
下载链接
链接失效反馈官方服务：
资源简介：
该数据集由来自SNOMED CT中已停用概念的术语对组成，这些术语对基于它们的关联被分类为在语义上相似。该数据集是自动创建的，根据莱文斯坦距离，它包含了从简单到困难的各种实例。该数据集规模宏大，包含了349,548个医疗概念，其任务是对医学术语相似性进行二分类。
This dataset consists of term pairs derived from retired concepts within SNOMED CT, which are classified as semantically similar based on their associated relationships. Automatically constructed, this dataset includes instances spanning from simple to difficult levels according to the Levenshtein distance. As a large-scale dataset containing 349,548 medical concepts, its core task is binary classification for medical term similarity.
提供机构：
Authors of the paper
