Electrical Engineering
      and Computer Sciences

Electrical Engineering and Computer Sciences


UC Berkeley


Research Projects

Predicting Key Users with Pre-Emergent Behavior

Albert Segars

Online communities share a number of qualities and often operate similarly to groups in shared physical locations [1]. Since an increasing amount of collaboration and learning is taking place in these online communities in today's society, it is becoming increasingly important that we explore the differences between groups that succeed and groups that fall apart in the online space. I am investigating whether a particular set of metrics can be used to recognize key members before they move into roles that are critical to maintaining the health of an online community. I believe that it will be possible to identify individuals who are likely to move into critical roles by analyzing community interaction information. Social scientists have outlined a number of roles that influence offline groups [2, 4] and some of these have already been identified in online communities [3, 6, 7], but it is unclear whether these individuals can be identified before achieving the role. My system's interfaces will be transparent to the users themselves but will be used to log user actions (where applicable, since the majority of this data is gathered from third parties). A wide variety of usage information can be monitored and recorded about individual behaviors, but I will be using publicly available data involving interactions between multiple users such as postings, responses to posts, and directed comments due to its ease of acquisition.

McKenna, K. Influences on the nature and functioning of online groups. In A. Barak (Ed.), Psychological aspects of cyberspace: Theory, research, applications (pp. 228-242). Cambridge, UK: Cambridge University Press.
Benne, K., Sheats, P. Functional roles of group members. The Journal of Social Issues, Volume 4. April 2010.