An Analysis of Participation Structure in Conversation Based on Interaction Corpus of Ubiquitous Sensor Data

We propose the application of ubiquitous sensor technology to analyze dynamic structure transitions and the exchange of non-verbal information in multi-party human-to-human conversational interactions in open situations. We present an analysis of conversational interactions that took place in an open interaction space of poster presentation sessions and were collected as an Interaction Corpus. Based on a model of conversational participation structure, we hypothesized that the audience diversity in conversations can be understood by observing the role of each person standing in front of a poster and an exhibitor. We argue that the conversational interactions observed in the Interaction Corpus can effectively be captured by the differences in the patterns of transition of the roles each participant plays in the conversational participation structure.