首页> 外国专利> KNOWLEDGE AMOUNT ESTIMATION INFORMATION GENERATING APPARATUS, AND KNOWLEDGE AMOUNT ESTIMATING APPARATUS, METHOD AND PROGRAM

KNOWLEDGE AMOUNT ESTIMATION INFORMATION GENERATING APPARATUS, AND KNOWLEDGE AMOUNT ESTIMATING APPARATUS, METHOD AND PROGRAM

机译:知识量估计信息生成设备,知识量估计设备,方法和程序

摘要

PROBLEM TO BE SOLVED: To perform estimation of a knowledge amount of an inquirer as to a topic, on which a conversation is held, on the basis of a conversation result with high accuracy.SOLUTION: A speech intention assessing unit 2 assesses, on the basis of text data TD of a speech sequence in a conversation between an inquirer and a respondent, a speech as to whether a speech intention is applicable to a question to an interlocutor or an explanation thereto; a feature quantity extracting unit 3 extracts an intention feature quantity indicative of a feature quantity as to a generation state of the speech intention in the TD, and extracts a total number of word types used by the inquirer from the TD as a used vocabulary feature quantity; a knowledge amount estimation information generating unit 4 generates estimation information used for estimation of a knowledge amount corresponding to the TD by using, as learning data, both of the feature quantities, and previously imparted knowledge amount information indicative of a knowledge amount assumed from the conversation in the TD with respect to the TD, from which both of the feature quantities are extracted; and a knowledge amount estimating unit 5 estimates a knowledge amount corresponding to the TD, which is an estimation object, on the basis of the intention feature quantity on the estimation object and the used vocabulary feature quantity on the estimation object, which are extracted from the TD serving as the estimation object, and the estimation information generated at the estimation information generating unit 4.
机译:解决的问题:基于高精度的对话结果,对询问者关于进行对话的话题的知识量进行估计。解决方案:语音意图评估单元2根据对话结果评估在询问者和应答者之间的对话中,基于语音序列的文本数据TD,关于语音意图是否适用于对话者的问题或对对话的解释的语音;特征量提取单元3提取指示关于TD中语音意图的生成状态的特征量的意图特征量,并且从TD中提取查询者使用的词类型的总数作为已使用词汇特征量;知识量估计信息生成单元4通过使用特征量和指示从会话中假定的知识量的先前赋予的知识量信息这两者作为学习数据来生成用于估计与TD相对应的知识量的估计信息。在相对于TD的TD中,从中提取两个特征量;知识量估计单元5基于从估计对象中提取的意图特征量和估计对象上的已使用词汇特征量,来估计与作为估计对象的TD相对应的知识量。用作估计对象的TD,以及在估计信息生成单元4处生成的估计信息。

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