The Construction And Attribute Reduction Of Fuzzy Concept Lattice

The theory of concept lattice, also called formal concept analysis, proposed by the German mathematician Wille R. in 1982, is one of the effective mathematical tools for conceptual data analysis and knowledge processing. It is perfect in theory, widely extendly in application since the 1990s. Nowadays, the theory of concept lattice has been widely applied to knowledge discover, information retrieval, data mining, machine learning, granular computing, etc.Formal concept analysis mainly focus on a classic formal context, but in real-life problems, the relation between objects and attributes is fuzzy and uncertain, the initial formal concept analysis has been difficult to express these fuzzy, uncertain information. Therefore, extends formal concept analysis via fuzzy logic reasoning or fuzzy set theory to fuzzy formal concept analysis, can solve these reality problems which are fuzzy and uncertain more confucive.This paper discusses variable precision fuzzy information granule based on fuzzy formal context, then methods of transforming variable precision fuzzy information granule are pre-sented; In addition, on the basis of fuzzy formal context, constructs fuzzy formal concept lattice, and formulate a attribute reduction method in fuzzy formal context via combining granular com-puting; Furthermore, from the perspective of rule acquisition, gives an approach to attribute redut in fuzzy decision formal context via discernibility matrices and discernibility function.