Expertise Location and Maintenance System in Online Help Communities

(*this reflects the project description on the SocialWorlds group website.)

Jun Zhang, Mark Ackerman, Lada Adamic, Kevin Nam | Publication: (UIST’07)

Keywords: expertise finding, organizational memory, collective intelligence, knowledge management, online coummunities

A standard problem in organizations and online communities, especially those devoted to help, is maintaining the expertise level required to answer questions, provide help, and maintain social engagement.   This is a complex problem, especially for large social settings.

QuME has been designed to tackle the expertise location and maintenance problem.  The QuME system facilitates large online help communities.  It does so by providing better mechanisms to automatically distribute tasks to people who want expertise and to those who are willing to help.

Overall, the QuME system will include:

  • An expertise profiling component:  This is the key mechanism to mine social characteristics and expertise topics from communication archives or online participation.  Our profiling component will handle both topic and expertise level.  This expertise profiler will automatically mine people’s previous answers, or extra resources if permitted, such as emails, reports, projects involved, etc. to find who knows what.
  • An expertise finding engine: This component will match those needing expertise (such as people asking questions or needing help) to those with expertise.  For an online community, we will build an expertise finder to augment the forum.  For an organization, we will construct an expertise finder to augment either an online community within that organization or the organization’s standard communication channels.
  • A social networking component: We will also build in the social network based expertise searching algorithms that we have previously studied [10, 12 above]. Thus, the system will not only reach people who registered in the online forum, but also people who are several network steps away. With the right incentive mechanism, this can greatly increase the participation of the online community.  This component also includes communication and transport mechanisms.
  • A social facilitation component:  QuME will include various mechanisms for inviting new participation, encouraging existing participants to continue, and facilitate those with greater expertise to participate without overloading them.  One of the findings from our research is that there is considerable value in having those with middle levels of expertise help those with lower levels; thus, we especially wish to encourage this middle range.  (This is also valuable from an organizational learning perspective.)  We will construct suitable prototypes of matching interfaces that provide this social facilitation as well as tools so people may manage their workload from their participation.  This is tightly connected to the expertise finding engine.