A Computing Model For Concept Fusing And Document Classification

Document classification is a challenging issue in knowledge management. Current methodologiessupervised by different point of views diversify their ways both in semantic representation and statisticalmeasurements. This paper proposes a novel approach to classify documents combining the characters ofsuch two methodologies. Document is divided in semantic segments. Formal Concept Analysis (FCA) isintroduced in defining the semantic interconnection from content segment of document and itscorresponding concept identifiers. The approach implements statistical estimate to construct weightedsemantic interconnection. Furthermore, the proposed model clusters and fuses concept identifiers toconstruct semantic overlay for specific document resources. Experiments of clustering and classifyingresearch papers show that the proposed approach is feasible and effective.