首页> 外国专利> DOCUMENT SET FEATURING METHOD, DOCUMENT SET RETRIEVAL METHOD USING THE SAME AND DEVICE THEREFOR

DOCUMENT SET FEATURING METHOD, DOCUMENT SET RETRIEVAL METHOD USING THE SAME AND DEVICE THEREFOR

机译:文档集特征化方法,使用相同方法和装置的文档集检索方法

摘要

PROBLEM TO BE SOLVED: To provide a document set featuring method, a document set retrieval method using the method and the device by which the similarity of document sets can surely be judged and similar document sets can be retrieved by turning vector data for which the features of documents for constituting the document set are elements to the feature amount of the document set.;SOLUTION: A tree t is constituted of the entire documents d11, d12, d13, d21, d22, d23, d31, d32 and d33 for constituting the document sets s1, s2 and s3 and the tree t constituted in such a manner is divided into some segments g1 and g2. Distributions for indicating how many documents are included in the divided respective segments g1 and g2 for the respective document sets s1, s2 and s3 are decided as the feature amounts s1 (1, 2), s2 (2, 1) and s3 (1, 2) of the respective document sets s1, s2 and s3.;COPYRIGHT: (C)2001,JPO
机译:解决的问题:为了提供一种文档集特征化方法,一种使用该方法和装置的文档集检索方法,通过该方法和装置可以确定文档集的相似性,并且可以通过旋转具有该特征的矢量数据来检索相似文档集。构成文档集的文档数量是文档集特征量的元素。解决方案:树t由构成文档集的整个文档d11,d12,d13,d21,d22,d23,d31,d32和d33组成。文档集s1,s2和s3以及以这种方式构成的树t被分成一些段g1和g2。将用于指示针对各个文档集s1,s2和s3的划分的各个段g1和g2中包括多少个文档的分布确定为特征量s1(1、2),s2(2、1)和s3(1, 2)各自的文档集s1,s2和s3 。;版权:(C)2001,JPO

著录项

  • 公开/公告号JP2001249951A

    专利类型

  • 公开/公告日2001-09-14

    原文格式PDF

  • 申请/专利权人 KDDI CORP;

    申请/专利号JP20000061096

  • 申请日2000-03-06

  • 分类号G06F17/30;

  • 国家 JP

  • 入库时间 2022-08-22 01:34:00

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