A data mining approach to identify obligation norms in agent societies

Abstract

Most works on norms have investigated how norms are regulated using institutional mechanisms. Very few works have focused on how an agent may infer the norms of a society without the norm being explicitly given to the agent. This paper describes how an agent can make use of the proposed norm identification architecture to identify norms. This paper explains how an agent using this architecture identifies one type of norm, an obligation norm. To this end, the paper proposes an Obligation Norm Inference (ONI) algorithm which makes use of association rule mining approach to identify obligation norms. © 2010 Springer-Verlag Berlin Heidelberg.