We try to use information extraction technology in some parts of the medical records and extract disease information to accumulate experience for extracting complete information from medical records.This paper attempts to use dictionary and rules to achieve the named entity recognition.Information extraction is based on shallow parsing and use pattern sentence matching method with the help of a 3 levels finite state automaton.%尝试将信息提取技术应用于病历的部分内容,提取其中的病症相关信息,为完整电子病历的信息提取累积经验.采用基于术语库和规则相结合的方法进行实体识别.信息提取基于浅层语法分析,通过句型模式匹配来实现.整个信息提取过程借助一个3层的有限状态自动机来完成.
展开▼