权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Research on generalizing a method for explaining information dynamically

推广动态解释信息方法的研究

基本信息

批准号：
15300041
负责人：
SHIMAZU Akira
金额：
$ 8.38万
依托单位：
JapanAdvanced Institute of Science and Technology
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (B)
财政年份：
2003
资助国家：
日本
起止时间：
2003 至 2006
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-15300041/
关键词：
dialogue dynamic explanation spoken language utterance dialogue structure 発話単位漸次的発語単位

项目摘要

In order to obtain an information processing model in human language communication, to establish a friendly human-machine interface technology, and to enhance a method for explaining information interactively, we have conducted a research such as recording and analysis explanation dialogues, modeling of an interactive explanation based on small grained utterance units, modeling of interpretation of recipient responses, extraction of local dialogue structures, extraction of discourse patterns which define a part of the interactive explanation model, and so on. We call an interactive explanation a dynamic explanation.In this research, we especially aim to study generality of the basic explanation model which we proposed based on analysis of traffic route explanation dialogues. From such a viewpoint, in order to analyze characteristics of dynamic explanation in spoken language and to generalize the model based on the analysis, we focused on recording, analyzing and modeling of other explanation dialogues in addition to the traffic route explanation. By this research, we mainly made a corpus of explanation dialogues in spoken language, analyzed the dynamic explanation functionally and semantically, made supporting tools for the analysis, proposed an generalized model for explaining information dynamically, and extracted functional patterns of local dialogue structures and semantic patterns of dynamic explanation.

为了获得人类语言交流中的信息处理模型，建立友好的人机界面技术，增强信息交互解释的方法，我们进行了解释对话的记录与分析、基于小粒度话语单元的交互解释建模、接受者响应的解释建模、局部对话结构的提取、话语模式的提取定义了交互解释模型的一部分，等等。我们称互动解释为动态解释。在本研究中，我们特别致力于研究基于交通路线解释对话分析的基本解释模型的通用性。从这个角度出发，为了分析口语动态解说的特点，并在分析的基础上归纳出模型，我们重点对交通路线解说之外的其他解说对话进行了记录、分析和建模。本研究主要构建口语解释对话语料库，对其动态解释进行功能和语义分析，并为分析提供支持工具，提出了信息动态解释的广义模型，提取了局部对话结构的功能模式和动态解释的语义模式。