Development of Argumentation Agent Based on Multimodal Information

2017 Fiscal Year Final Research Report

• Project/Area Number: 15H02746
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Tokyo Institute of Technology
• Principal Investigator: Nitta Katsumi 東京工業大学, 情報理工学院, 教授 (60293073)
• Co-Investigator(Kenkyū-buntansha): 岡田 将吾 北陸先端科学技術大学院大学, 情報科学系, 准教授 (00512261) ; 平田 勇人 朝日大学, 法学部, 教授 (90189837)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 人工知能 / 対話エージェント / マルチモーダル情報 / 議論演習 / 数理議論学 / 議論スキル

Outline of Final Research Achievements

To reduce the burden of managing argumentation training of a supervisor of a law school, we developed an argumentation agent which argues with students concerning a given case and evaluates argumentation skills. The argumentation agent understands utterances of students, judges the agent is getting advantage or not, estimates emotinal status of students, then selects the proper response according to an argumentation strategy. This agent is developed on the humanoid robot, Pepper. After the argumentation, this agent evaluates various argumentation skills such as logicalness, persuasiveness, riposte, keeping working relation and so on, and then shows these scores to the students.

Free Research Field

人工知能