Towards Cooperative Predictive Data Mining in Competitive Environments

Abstract. We study the problem of predictive data mining in the competitive multi-agent setting, in which each agent is assumed to have some partial knowledge needed for correctly classifying a set of unlabelled examples. The agents are self-interested and therefore need to reason about the trade-offs between increasing their classification accuracy by collaborating with other agents and disclosing their private classification knowledge to other agents through such collaboration. We analyze the problem and propose a set of components which can enable cooperation in this otherwise competitive problem. These components include measures for quantifying private knowledge disclosure, data-mining models suitable for multi-agent predictive data mining, and a set of strategies by which agents can improve their classification accuracy through collaboration. The overall framework and its individual components are validated on a synthetic experimental domain.

• Type: Conference
• Year: 2009
• Where: ADMI
• Authors: Viliam Lisý, Michal Jakob, Petr Benda, Stepán Urban, Michal Pechoucek
• Keywords: ADMI 2009, Classification Accuracy, Competitive Multi-agent Setting, Data Mining, Predictive Data Mining