Multimodal Analysis and Concern Alignment Model for Mutual Trust Formation in Conversations

2015 Fiscal Year Final Research Report

• Project/Area Number: 24300061
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Future University-Hakodate
• Principal Investigator: KATAGIRI YASUHIRO 公立はこだて未来大学, システム情報科学部, 教授 (60374097)
• Co-Investigator(Kenkyū-buntansha): ISHIZAKI MASATO 東京大学, 大学院情報学環, 教授 (30303340) ; DEN YASUHARU 千葉大学, 文学部, 教授 (70291458) ; TAKANASHI KATSUYA 京都大学, 学術情報メディアセンター, 研究員 (30423049) ; ENOMOTO MIKA 東京工科大学, メディア学部, 講師 (10454141)
• Co-Investigator(Renkei-kenkyūsha): SUMI YASUYUKI 公立はこだて未来大学, システム情報科学部, 教授 (30362578) ; OKADA SYOGO 東京工業大学, 総合理工学研究科, 助教 (00512261) ; NAMBU MISAKO 公立はこだて未来大学, システム情報科学部, 准教授 (10404807)
• Project Period (FY): 2012-04-01 – 2016-03-31
• Keywords: 自然言語処理 / 会話処理 / 相互信頼感

Outline of Final Research Achievements

This research was aimed at establishing a computational model of the processes of consensus building and mutual trust formation/maintenance through conversational interactions. Real-life medical communication and business communication dialogue corpus data were analyzed in terms of the notion of 'concern alignment' we have proposed. We demonstrated that the concern alignment model of conversation can provide a useful framework to describe strategic interactions for consensus building as well as underlying value sharing for trust formation by capturing macro-level discourse structures manifesting and supporting the processes. A simple game-theoretical model of multi-issue negotiation was also proposed as a basis for computational modeling of the processes of concern alignment.

Free Research Field

知能情報学