首页> 外国专利> NAMED-ENTITY RECOGNITION METHOD AND APPARATUS, COMPUTER DEVICE AND READABLE STORAGE MEDIUM

NAMED-ENTITY RECOGNITION METHOD AND APPARATUS, COMPUTER DEVICE AND READABLE STORAGE MEDIUM

机译:命名实体识别方法和装置,计算机设备和可读存储介质

摘要

A named-entity recognition method and apparatus, a computer device and a readable storage medium. Said method comprises: acquiring a medical text, and preprocessing the medical text, so as to obtain a text to be processed (S100); on the basis of a preset dictionary, performing microbial entity extraction on the text to be processed, so as to obtain a target entity (S200); generating a plurality of candidate abbreviation entities according to a first preset rule and the target entity, and performing screening, by using a first model, from the candidate abbreviation entities, so as to obtain the candidate abbreviation entity corresponding to the entity and take same as a target abbreviation entity (S300); generating a plurality of candidate supplementary entities according to a second preset rule and the target entity, and screening the candidate supplementary entities by using a second model, so as to obtain a target supplementary entity (S400); and generating target data on the basis of the target entity, the target abbreviation entity and the target supplementary entity (S500). The present invention solves the technical problem of relatively low accuracy caused by a dictionary matching-based entity extraction method being unable to take entities having abbreviations or specific information into consideration.
机译:命名实体识别方法和装置,计算机设备和可读存储介质。所述方法包括:获取医疗文本,并预处理医疗文本,以便获取要处理的文本(S100);在预设字典的基础上,对要处理的文本执行微生物实体提取,以便获得目标实体(S200);根据第一预设规则和目标实体生成多个候选缩写实体,并通过使用第一模型从候选缩写实体执行筛选,以便获得与实体对应的候选缩写实体并采用相同目标缩写实体(S300);根据第二预设规则和目标实体生成多个候选补充实体,并通过使用第二模型来筛选候选补充实体,以便获得目标补充实体(S400);并基于目标实体,目标缩写实体和目标补充实体生成目标数据(S500)。本发明解决了由基于词典匹配的实体提取方法引起的相对低的精度的技术问题,不能考虑具有缩写或特定信息的实体。

著录项

  • 公开/公告号WO2021179708A1

    专利类型

  • 公开/公告日2021-09-16

    原文格式PDF

  • 申请/专利权人 PING AN TECHNOLOGY (SHENZHEN) CO. LTD.;

    申请/专利号WO2020CN134882

  • 发明设计人 GU DAZHONG;ZHANG SHENG;

    申请日2020-12-09

  • 分类号G06F40/295;G06F16/33;

  • 国家 CN

  • 入库时间 2022-08-24 21:08:19

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号