首页> 外国专利> keyword extraction method and apparatus for science document

keyword extraction method and apparatus for science document

机译:科技文献关键词提取方法及装置

摘要

According to an embodiment of the present invention, a method of extracting key words of a scientific document may include: converting a scientific document into a morphologically analyzed scientific document consisting of words analyzed by nouns, adjectives, and verbs through morphological analysis; Constructing a document graph by setting a word of the scientific document as a vertex and giving an edge indicating a relationship between the word and the word; Calculating a importance score for the vertices in the document graph; Detecting location information of key phrase candidates and extracted key phrase candidates in the morphologically analyzed scientific document; Calculating scores of candidate key phrases according to scores of words included in key phrase candidates and lengths of key phrase candidates for each of the key phrase candidates; Reranking the ranking of the key phrase candidates by changing the score of each key phrase candidate according to the position information of the key phrase candidate; And determining a predetermined number of upper key phrase candidates among key phrase candidates that have been reranked as key words of the scientific document.
机译:根据本发明的实施例,一种提取科学文献的关键词的方法可以包括:将科学文献转换成由名词,形容词和动词通过形态分析所分析的词组成的形态分析的科学文献;通过将科学文献的单词设置为顶点并给出表示单词与单词之间的关系的边来构造文档图;计算文档图中顶点的重要性得分;在形态分析的科学文献中检测关键词候选者的位置信息和提取的关键词候选者;根据关键字候选中包含的单词的分数和每个关键字候选的关键字候选的长度计算候选关键字的分数;通过根据关键词候选者的位置信息改变每个关键词候选者的得分来重新排列关键词候选者的排名;并且在已经被重新排名为科学文献的关键词的关键词候选中确定预定数量的上部关键词候选。

著录项

  • 公开/公告号KR102017227B1

    专利类型

  • 公开/公告日2019-09-02

    原文格式PDF

  • 申请/专利权人 서강대학교산학협력단;

    申请/专利号KR20170145461

  • 发明设计人 서정연;고영중;염홍선;

    申请日2017-11-02

  • 分类号G06F17/27;G06F17/10;

  • 国家 KR

  • 入库时间 2022-08-21 11:47:49

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