Hidden Interaction Pattern Discovery Of Human Interaction In Meetings

Abstract: In this paper, we propose a mining method to extract frequent patterns of human interaction based on the captured content of face-to-face meetings. Human interaction, such as proposing an idea, giving comments and expressing a positive and negative opinion, indicate user topic or role in a discussion. Human interaction flow in a discussion session is represented as a tree. We developed several applications based on the discovered patterns. We explore embedded tree mining for hidden interaction pattern discovery. Embedded sub trees are a generalization of induced sub tree, which allow not only direct parent child branches, but ancestor-descendant branches. The current meetings are all task oriented. It is valuable to capture various categories of meetings for analysis such as panel, debate, interview etc. There would be some differences in the frequent interaction patterns for different meeting styles.

Keywords: patterns, human interaction, tree mining, pattern discovery, embedded.