Fuzzy-weighted distance and its applications in pattern recognitionand classification

摘要

The fuzzy-weighted distance presented shows that every feature ofevery sample has a different effect on distance in pattern recognitionand classification. The difference is described by using the concept offeature odds defined by the authors, which is derived from fuzzy settheory and is determined by human thinking in object recognition. Thereasonableness of the proposed distance is discussed. Experimentalresults are presented that show the advantages of this distance