首页> 外国专利> SEMANTIC DISAMBIGUATION METHOD OF CHINESE PLACE NAMES BASED ON ENCYCLOPEDIA KNOWLEDGE BASE AND WORD VECTORS

SEMANTIC DISAMBIGUATION METHOD OF CHINESE PLACE NAMES BASED ON ENCYCLOPEDIA KNOWLEDGE BASE AND WORD VECTORS

机译:基于百科全书知识库和文字向量的中文名称的语义消歧方法

摘要

Provided is a semantic disambiguation method of Chinese place names based on encyclopedia knowledge base and word vectors, comprising: generating a pre-training word vector model; generating a place name recognition model based on Bi-LSTM and CRF integration; performing place name recognition and word vector representation in a text where the place name to be disambiguated is located; obtaining sense name of the ambiguous place name and sense content thereof; performing place name recognition and word vector representation in sense content of the ambiguous place name; performing similarity calculation on the place name list word vector in the text where the ambiguous place name is located and the place name list word vector in the sense content of the ambiguous place name; performing sense frequency statistics of ambiguous place names; performing comprehensive evaluation of the geographic position which the ambiguous place name most possibly refers to. The above method can effectively solve the problems that the same place name may correspond to a plurality of geographic positions and the same position may have a plurality of description names in the position service, is suitable for the situation that the place names in text description are distributed discretely and sparsely under the historical background of the Internet+, and meets the urgent requirements for intelligent position services, holographic position maps and the like.
机译:提供了基于百科全书知识库和文字矢量的中文名称的语义消除歧义方法,包括:生成预训练的单词矢量模型;生成基于Bi-LSTM和CRF集成的地方识别模型;在文本中执行位置名称识别和字向量表示,其中要消除的位置名称所在;获取含糊不清的地方名称和感应内容的感应名称;在含糊不清地名的感觉内容中执行地名识别和字矢量表示;在歧义地点名称所在的文本中的地点名单中执行相似性计算,并且在模糊的地方名称的感测内容中的位置名单列表字向量;执行含糊不清地名的感测频率统计;对地理位置的全面评估,模糊的地方最可能是指的。上述方法可以有效地解决了相同的位置可以对应于多个地理位置的问题,并且相同的位置可以具有位置服务中的多个描述名称,适用于文本描述中的地点名称的情况在互联网+的历史背景下离心和稀疏地分发,符合智能职位服务,全息位置图等的迫切要求。

著录项

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