Online communities generate major economic value and form pivotal parts of corporate expertise management, marketing, product support, CRM, product innovation and advertising. Communities can exceed millions of users and infrastructures must support hundreds of millions discussion threads that link together billions of posts. ROBUST is targeted at developing methods to understand and manage the business, social and economic objectives of the users, providers and hosts and to meet the challenges of scale and growth in large communities. Hence, the objectives of ROBUST are to find solutions for community risk management, large scale data management, models of community polity and politics, community simulation and community data analysis. The objectives of ROBUST fall into the following categories:

  • Risk Management: The identification and modelling of risks andopportunities in online communities will support the understanding andmanagement of these communities. In particular it will enablestakeholders to identify threats, support their decision making processand lead them in choosing proactively measures to counter risks or seize opportunities.
  • Community Data Management: The volume of data created in online communities in the form of texts, the interaction between users or simply the interaction of users with the system itself demands for new technologies for large scale data management and processing.
  • Community Polity and Politics: Understanding the behaviour and needs of users on a micro level requires detailed user models. This allows to classify users based on behaviour patterns and determine the role they play in a community, what is their status and what motivates their actions.
  • Community Simulation: A model on a macro-level captures the dynamics of entire communities and their development. Understanding the effects of policies on a community can help to forecast the way the community is evolving and in which direction it is heading.
  • Community Analysis: The ability to detect communities, find thetopics they are dealing with and to recognize patterns in massivecommunity data complements the other objectives.


Funded by