Published Articles >> Table of Contents >> Abstract

2001 IEEE International Conference on Multimedia and Expo (ICME'01)   p. 112
ZINGER: CONVERSATION SUPPORT BASED ON INTEREST SIMILARITY
Noriyuki Kurabayashi, Research Laboratories Seika-cho, Soraku-gun
Tatsuya Yamazaki, Research Laboratory Informatics, Seika-cho, Soraku-gun
Kazuo Hasuika, Research Laboratories Seika-cho, Soraku-gun
Taiichi Yuasa, Informatics, Kyoto University

Full Article Text:

DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ICME.2001.1237750
Send link to a friend

Abstract
In this paper, we introduce a conversation support agent called Zinger. Zinger acquires the current interest of a participant, identifies a partner who has a similar interest, and recommends several topics related to the interest. This recommendation is expected to help the participant establish common ground in conversing with the partner. The participant can get valuable information with a successful conversation. Experimental results show that Zinger could promote utterances of 35% subjects. In addition, factors affecting the participant's utterances are examined by multiple regression analysis. The results suggest that providing helpful information and appropriate presentations can contribute to better utterances.
Back to Top Additional Information
Citation:  Noriyuki Kurabayashi, Tatsuya Yamazaki, Kazuo Hasuika, Taiichi Yuasa, "ZINGER: CONVERSATION SUPPORT BASED ON INTEREST SIMILARITY," icme, p. 112,  2001 IEEE International Conference on Multimedia and Expo (ICME'01),  2001