Solving complex problems by multi-agent systems in distributed environments requires new approximate reasoning methods based on new computing paradigms. One such recently emerging computing paradigm is Granular Computing. Granular computations are performed on information granules representing vague and complex concepts delivered by agents engaged in tasks such as knowledge representation, communication with other agents, and reasoning. In this paper, we present an outline of foundations for information granule calculi and methods for inducing relevant information granule constructions from data and background knowledge. These constructions can be interpreted as approximate reasoning schemes. The proposed methodology of approximate reasoning has been developed for solving complex problems in areas such as identification of objects by autonomous systems, web mining or sensor fusion.
展开▼