In the research field of human-computer interaction, there are many approaches to predicting interactive roles, e.g., conversational dominance or active participation. Although interactive roles have been predicted for entire tasks, little attention has been given to evaluating how such roles are reorganized during a task. This paper explains how to construct a model for predicting emergent division of labor and the reorganization of labor in multiparty collaboration using verbal and nonverbal cues. To build the model, we adopted stepwise multiple-regression analysis, which is a type of statistical model analysis, using both behavioral data and third-party evaluations. We confirmed useful verbal and non-verbal parameters for predicting interactive roles and their reorganization through this model. © 2013 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Suzuki, N., Kamiya, T., Umata, I., Ito, S., Iwasawa, S., Sakata, M., & Shimohara, K. (2013). Detection of division of labor in multiparty collaboration. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8018 LNCS, pp. 362–371). https://doi.org/10.1007/978-3-642-39226-9_40
Mendeley helps you to discover research relevant for your work.