Publication Type:Journal Article
Source:User Modeling and User-Adapted Interaction: The Journal of Personalization Research, Springer, Volume Accepted (2013)
Individuals participating in technologically mediated forms of organization have difficulty recognizing when groups emerge, and how the groups they take part in change over time. One way of increasing this awareness is to increase the amount of contextual information available to participants. This information will enable group members to adapt to the context of group membership. We demonstrate how traces of technologically mediated interactions between group members can be can used to describe the emergence of group context. Such virtual groups arise as individual contexts overlap, merge and are combined into a common work context for the group, which - in turn - informs the individual activities and forms the substrate for collaboration and coordination. This paper has two parts. First, it describes five different cases where we have analyzed how to construct group and task context by combining electronic trace data analysis with other methods. The cases are descriptive in nature; with each one describing how context is discoverable from electronic trace data, how that discovery could be automated, and what additional data, if captured, would enable automated context adaptation. The five case studies include domains of software engineering, online learning, disaster relief and public political discourse. Second, in our discussion we describe methods, data management strategies and technical architecture for identifying emergent small groups from available electronic trace data. We argue that context adaptive systems must move fluidly between individual and group context adaptation.