Knowledge Acquisition via Non-monotonic Reasoning in Distributed Heterogeneous Environments

The role of data and knowledge exchange is becoming increasingly important. The approach of DACMAS [1] proposes a quite general modeling of Multi-Agent Systems (MAS), including data representation in a MAS via DRL-Lite Ontologies. Yet, data/knowledge acquisition from heterogeneous sources which are not agents and which are external to the MAS is not provided. In the Knowledge Representation and Reasoning field, this topic is coped with by mMCSs (Managed Multi-Context Systems). In this paper, we propose an integration of the two approaches into DACMACSs. The aim is to obtain an enhanced integrated flexible framework where non-monotonicity is present: in the modalities for defining knowledge acquisition; in the conditions for triggering the acquisition and for knowledge exploitation.