Presentation 2012-06-29
Hot spot detection in multi-party conversation using laughing feature
Kazutaka SHIMADA, Akihiro KUSUMOTO, Takahiko YOKOYAMA, Tsutomu ENDO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a method for hot spot detection in multi-party conversation. Detecting hot spots in conversation helps to recognize user's minds, and leads to the improvement of dialogue understanding. We focus on laughing of each user. We divide laughing situations into "internal" and "external". The "internal" and the "external" denote unprompted laughing and laughing that is caused by actions or utterances from other persons, respectively. Besides, we classify the power of laughing into "smile", "laugh" and "burst". We generate a classifier with these laughing features and basic features such as bag-of-words. We evaluate the proposed method with 10 dialogue sets. The experimental result shows the effectiveness of the proposed method with laughing features.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Multi-party conversation / Hot spot detection / Laughing feature
Paper # NLC2012-7,PRMU2012-27
Date of Issue

Conference Information
Committee NLC
Conference Date 2012/6/22(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) Hot spot detection in multi-party conversation using laughing feature
Sub Title (in English)
Keyword(1) Multi-party conversation
Keyword(2) Hot spot detection
Keyword(3) Laughing feature
1st Author's Name Kazutaka SHIMADA
1st Author's Affiliation Department of Artificial Intelligence, Kyushu Institute of Technology()
2nd Author's Name Akihiro KUSUMOTO
2nd Author's Affiliation Department of Artificial Intelligence, Kyushu Institute of Technology
3rd Author's Name Takahiko YOKOYAMA
3rd Author's Affiliation Department of Artificial Intelligence, Kyushu Institute of Technology
4th Author's Name Tsutomu ENDO
4th Author's Affiliation Department of Artificial Intelligence, Kyushu Institute of Technology
Date 2012-06-29
Paper # NLC2012-7,PRMU2012-27
Volume (vol) vol.112
Number (no) 110
Page pp.pp.-
#Pages 6
Date of Issue