Poster Submissions

Poster 1

Remote Meeting Tension Relief through Transformation of Interviewer's Face [poster abstract]

Ziting Gong and Hideaki Kanai, Japan Advanced Institute of Science and Technology, Japan

Abstract In this study, we propose a method for transforming an interviewer's face into that of an acquaintance in a remote meeting. We examine whether this method could reduce the tension of the speaker through the experiment. The results of a subjective questionnaire and a facial skin temperature measurement showed that the speakers' tension in the case of facing the scenes of an acquaintance was lower than that in the case of facing the scenes of a stranger. Therefore, this demonstrates that the proposed method is e ective in relieving the speaker's tension during a remote meeting.

 

Poster 2

Survey of User Trust for the Video-Recommendation Function in YouTube [poster abstract]

Reika Miwa*, Aika Tsuchida*, Yoshinori Hijikata*, Masahiro Hamasaki** and Masataka Goto**

*Kwansei Gakuin University, Nishinomiya, Hyogo, Japan
** National Institute of Advanced Industrial Science and Technology (AIST), Ibaraki, Japan

Abstract In this paper, we investigate the users’ trust against the video recommendation function of YouTube, the most popular video sharing service. The perceived usefulness and the perceived ease of use, which are used in TAM (a model of the user’s acceptance of new products implemented with novel technologies), are also investigated in order to clarify which factor of trust is correlated. We conducted a crowdsourcing survey of 631 users and found that the perceived usefulness and the perceived ease of use are highly correlated with the intention to reuse the recommendation function and the degree of overall trust on it.

 

Poster 3 

Improving Cultural Difference Detection System Using Confidence Value [poster abstract]

Mondheera Pituxcoosuvarn and Yohei Murakami, Faculty of Information Science and Engineering, Ritsumeikan University, Japan

Abstract The Cultural Difference Detection(CDD) concept was introduced to enhance machine translation(MT) utility by predicting possible misunderstanding. CDD compares images linked to a concept in two languages with the assumption that low image similarity may indicate possible cultural misunderstanding. This paper extends CDD by roposing a method to calculate confidence value that assess the accuracy of CDD. Confidence value can be used to decide whether to warn the user of a possible misunderstanding when a suspect word is used in a multilingual chat system.