利用GATE的XML配置文件实现病历短语抽取的机器学习方法  

Machine Learning Method to Realize Medical Record Phrase Extraction via Using the XML Configuration File of the GATE

作　　者：倪晓华[1] NI Xiao-hua(Department of Information, the SecondAffiliated Hospital of Nanjing MedicalUniversity, Nanjing Jiangsu 210011,China)

机构地区：[1]南京医科大学第二附属医院信息科,江苏南京210011

出　　处：《中国医疗设备》2017年第7期124-125,133,共3页China Medical Devices

摘　　要：本文利用文本工程通用框架软件的XML配置文件,来指定所学文档使用的特征参数、学习算法,实现文本病历医学短语抽取的机器学习。结果计算机能很方便的在大段病程资料中快速自动获取医生所需的医学短语信息。本学习算法具有较好的实用性,达到了预期要求。Based on XML configuration files of general architecture for text engineering,we specifiedcharacteristics and learning algorithm of the documents,and realized machine learning of text recordsphrase extraction.The result was that computer could automatically obtain the phrases that doctorrequired from the long course information quickly.This learning algorithm has good practicability andmeets the expected demand.

关 键 词：电子病历 机器学习 通用框架软件 支持向量机 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]