TCM-SD 中医领域辨证数据集

TCM-SD为首个公开、现实场景收集的中医辨证数据集，旨在利用自然语言处理技术解决中医辨证问题，探究中医辨证论治理论背后潜藏的科学依据。
相关论文“TCM-SD: A Benchmark for Probing Syndrome Differentiation via Natural Language Processing”发表在CCL2022会议上。

TCM-SD is the first publicly available TCM syndrome differentiation dataset collected from real-world scenarios. It aims to leverage natural language processing (NLP) technologies to address the challenges of TCM syndrome differentiation and explore the underlying scientific basis behind the theory of TCM syndrome differentiation and treatment. The associated paper titled "TCM-SD: A Benchmark for Probing Syndrome Differentiation via Natural Language Processing" was published at the CCL 2022 conference.

搜集汇总

背景与挑战

背景概述

TCM-SD是首个公开、基于现实场景收集的中医辨证数据集，包含54,152条临床记录和148种证型，旨在通过自然语言处理技术将中医辨证任务转化为计算机可处理的形式，即根据病人描述预测证型。该数据集遵循CC BY-NC-SA 4.0协议，仅用于学术研究，并经过脱敏处理，以促进中医领域的智能化研究。