Comprehend Medical:一个命名实体识别与关系抽取网络服务
摘要
Comprehend Medical 是一个无状态且符合《健康保险流通与责任法案》(HIPAA) 要求的命名实体识别 (NER) 与关系抽取 (RE) 服务,在亚马逊网络服务 (AWS) 下启动,并使用最先进的深度学习模型进行训练。与许多现有的开源工具不同,Comprehend Medical 具有可扩展性,且无需陡峭的学习曲线、依赖项、流水线配置或安装。目前,Comprehend Medical 在五个医学类别中执行命名实体识别:解剖、医疗状况、药物、受保护健康信息 (PHI) 以及治疗、测试与程序 (TTP)。此外,该服务还为检测到的实体提供关系抽取,以及以特征形式呈现的上下文信息,如否定和时序性。Comprehend Medical 提供两个应用程序编程接口 (API):1) NERe API,返回所有提取的命名实体、其特征及它们之间的关系;2) PHId API,仅返回文本中包含的受保护健康信息。此外,Comprehend Medical 可通过 AWS 控制台、Java 和 Python 软件开发工具包 (SDK) 访问,使非开发者和开发者都能更轻松地使用。
引用
@article{arxiv.1910.07419,
title = {Comprehend Medical: a Named Entity Recognition and Relationship Extraction Web Service},
author = {Parminder Bhatia and Busra Celikkaya and Mohammed Khalilia and Selvan Senthivel},
journal= {arXiv preprint arXiv:1910.07419},
year = {2019}
}
备注
ICMLA 2019. $\copyright$ 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses in any current or future media including reprinting/republishing this material for advertising, promotional purposes, creating new collective works, for resale or redistribution to servers or lists, reuse of any copyrighted component of this work in other works. arXiv admin note: text overlap with arXiv:1812.05270