Presentation 2007/7/17
A Study on Modeling Technique for Vague Expressions of Human Motion in Spoken Dialogue Agents
Makoto TACHIBANA, Keigo KAWASHIMA, Takao KOBAYASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes a statistical modeling technique for appropriately interpreting vague expressions that are appeared in human-computer interaction with spoken dialogue agents. Through human-human interactions, we collect many examples of human motion and their environments that are related to command words including vague expressions. In the statistical modeling of the vague expressions, the distributions which do not have enough training data are shared using decision-tree based context clustering technique. Prom experimental results, we show that the performance of the human-computer interaction using the proposed model is very similar to that of the human-human interaction. Furthermore, we apply reinforced learning to improve the model and respond to unknown command words.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) spoken dialogue agents / vague expressions / statistical model / decision tree / reinforced learning
Paper # NLC2007-4
Date of Issue

Conference Information
Committee NLC
Conference Date 2007/7/17(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Study on Modeling Technique for Vague Expressions of Human Motion in Spoken Dialogue Agents
Sub Title (in English)
Keyword(1) spoken dialogue agents
Keyword(2) vague expressions
Keyword(3) statistical model
Keyword(4) decision tree
Keyword(5) reinforced learning
1st Author's Name Makoto TACHIBANA
1st Author's Affiliation Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology()
2nd Author's Name Keigo KAWASHIMA
2nd Author's Affiliation Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology
3rd Author's Name Takao KOBAYASHI
3rd Author's Affiliation Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology
Date 2007/7/17
Paper # NLC2007-4
Volume (vol) vol.107
Number (no) 158
Page pp.pp.-
#Pages 6
Date of Issue