首页> 外国专利> AN AUTOMATIC CLUSTERING METHOD OF WORD SENSES USING A WORD RELATION GRAPH

AN AUTOMATIC CLUSTERING METHOD OF WORD SENSES USING A WORD RELATION GRAPH

机译:基于词关系图的词义自动聚类方法

摘要

PURPOSE: An automatic population technique of the lexeme of using the word linkage graph. Linkage words are constituted with words used in the actual data. It includes in the lexeme population to the new coined word or the foreign language. CONSTITUTION: In the same time prevalence table of the term frequency table and two words is the web document, noun is extracted. The mutual information between the different two words is calculated. The word linkage graph uses the linkage word satisfying the mutual information of the critical over and connects information. The word linkage graph smalls and it partitions to lower part populations of many numbers. The population result it about the mutual information threshold is selected by using the population evaluation index. The keyword which most strongly shows the meaning of population is selected among linkage words.
机译:目的:使用词链接图的词素的自动填充技术。链接词由实际数据中使用的词构成。它在词素族中包括新造词或外语。构成:在术语频率表的同时流行表中,两个单词是网络文档,名词被提取。计算两个不同单词之间的相互信息。单词链接图使用满足临界条件的相互信息的链接单词并连接信息。单词链接图很小,它划分为许多数字的较低部分人口。使用总体评估指标选择关于互信息阈值的总体结果。在链接词中选择最能显示人口意义的关键词。

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