首页> 外国专利> COMPOUND NOUN EXTRACTION DEVICE

COMPOUND NOUN EXTRACTION DEVICE

机译:复合名词提取装置

摘要

PPROBLEM TO BE SOLVED: To provide a compound noun extraction device allowing extraction of a proper compound noun without describing a compound noun list or a detailed rule in advance. PSOLUTION: The compound noun extraction device morpheme-analyzes document data, thereafter refers to a speech part connection rule by speech part information of a morpheme, and obtains compound noun candidate data 150 with continuous morphemes as a compound noun candidate when the continuous morphemes fit the connection rule. The compound noun extraction device obtains a forward score of the head morpheme and a rearward score of the final morpheme in reference to character string frequency data about each head morpheme and each final morpheme constituting the compound noun candidate, and extracts a character string from the head morpheme to the final morpheme as the compound noun when both the scores are larger than a score set value. PCOPYRIGHT: (C)2011,JPO&INPIT
机译:

要解决的问题:提供一种复合名词提取装置,其允许提取适当的复合名词而无需事先描述复合名词列表或详细规则。

解决方案:复合名词提取器对文档数据进行词素分析,然后通过词素的语音部分信息参照语音部分连接规则,获得连续词素作为复合名词候选词的复合名词候选数据150。语素符合连接规则。复合名词提取装置参照构成复合名词候选者的每个头部语素和每个最终语素的字符串频率数据,获得头部语素的前向得分和最终语素的后向得分,并从头部中提取字符串当两个分数均大于分数设定值时,将词素变为最终词素,作为复合名词。

版权:(C)2011,日本特许厅&INPIT

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