The Research Of A Block Lattice Algorithm Generating Fuzzy Concept Lattices By Similarity

Concept lattice theory of formal concept analysis is a core data structure, it has been widely used in many fields, such as knowledge discovery and information retrieval, and so on. The general form of conceptual analysis is based on the exact form of the background, but in our real world, due to the complexity and diversity of things, the people understand things, the definition of the concept is difficult to express with accurate data. So, in order to make the formal concept analysis be applied to a wide range of various research areas, there were some scholars combines fuzzy set theory, fuzzy logic and classical concept lattice theory (accurate concept lattice theory), to propose the fuzzy formal concept analysis and fuzzy concept lattice theory of this new research directions, while the problem of fuzzy concept lattice structure has become a primary subject to these researcher.According to the characterized of the concept lattice data structure, at present, the problems of how to deal with large-scale fuzzy formal context, and how to construct a large-scale fuzzy concept lattice has become one of the more major research subject. The German scholar Radim Belohlavek and the other men, in order to solve the problems of large-scale data generation concept lattice, simplifying large fuzzy concept lattice and the rapid generation of fuzzy concept lattices, propose a series of theories and algorithms. Which refers to the structure factor of the concept lattice, but it did not give a complete lattice construction algorithm factor. In order to improve and accomplish the fuzzy concept lattice factor lattice structure, this paper proposes a fuzzy concept factor lattice algorithm based on the fuzzy galois connection by similarity.First, the text of the relevant background fuzzy concept lattice and fuzzy concept lattice of related concepts were introduced. For related structures algorithms have been proposed fuzzy concept lattices are described and summarized, noting the current fuzzy concept lattice construction algorithm for the problems, and then put forward the thesis research.For purpose of accomplishing the construction of the factor lattice, this paper developed a factor lattice algorithm of the fuzzy concept lattice based on the fuzzy galois connection by similarity (Factor Lattice Algorithm, FLA). This algorithm, according to the threshold which is made by the users, provides a processing batch method to structure a complete fuzzy factor lattice, based on the supremum and infimum of the fuzzy concept block lattice and the generating factors to get the down-neighbor concept sets, by a top-down way, and the algorithm also gives a way to structure a local complete factor lattice. The theoretical analysis and experimental results show that this algorithm will structure a complete fuzzy concept factor lattice.