Title :
A new method for fuzzy pattern classification based on measures of similarity
Author :
Flokos, Ioannis I.
Author_Institution :
Dept. of Electr. Eng., Ruhr-Univ., Bochum, Germany
Abstract :
This paper presents a new method for fuzzy pattern classification. Its main difference from other methods is the introduction of a measure of similarity between the patterns and the point in the n-dimensional space which is to be classified. Furthermore this method is robust against noise corrupted patterns, which is an important aspect in many pattern classification problems
Keywords :
fuzzy set theory; pattern classification; fuzzy pattern classification; measure of similarity; noise corrupted patterns; pattern classification; robust; Bridges; Concrete; Engineering in medicine and biology; Euclidean distance; Fuzzy sets; Noise robustness; Pattern classification; Physics;
Conference_Titel :
Fuzzy Systems Symposium, 1996. Soft Computing in Intelligent Systems and Information Processing., Proceedings of the 1996 Asian
Conference_Location :
Kenting
Print_ISBN :
0-7803-3687-9
DOI :
10.1109/AFSS.1996.583544
