Incremental learning of concept cluster on background net for fuzzy granulation of document representation

Abstract

Having a term serving as symbol, capturing the meaning hidden behind the symbol becomes an important task affecting the understanding of the symbol. Concept cluster is proposed in this article for representing the contextual information of a term. Background net previously proposed by the same group of authors is a novel approach for capturing and representing background information as a knowledge background accumulated through incremental learning on articles. Through fuzzy concept clusters captured on background net, a document can be represented in the form of fuzzy granular. Based on the fuzzy concept clustering, algorithms for learning and categorization for text cate-gorization are also proposed. © 2012 ICIC International.