Title :
A fuzzy Min-Max neural network classifier based on centroid
Author :
Liu Jinhai ; He Xin ; Yang Jun
Author_Institution :
Sch. of Inf. Sci. & Eng., Northeastern Univ., Shenyang, China
Abstract :
Based on the analysis of the parameters of fuzzy Min-Max neural network, a new classification method of fuzzy Min-Max neural network for real data pattern is proposed. A new membership function is designed which control the descend of membership degree by the distance of the geometric center of hyperbox and Centroid, and the membership function can also adjust the boundary of hyperbox. The influence of parameters of new membership function is discussed by 1-D samples. The availability of the new method is validated by simulation of IRIS dataset.
Keywords :
data analysis; fuzzy neural nets; fuzzy set theory; minimax techniques; pattern classification; centroid; classification method; data pattern; fuzzy min-max neural network classifier; geometric center; hyperbox boundary; membership degree; membership function; parameter analysis; Biological neural networks; Cybernetics; Electronic mail; Iris; Neurons; Silicon compounds; Centroid; Classification; Fuzzy Min-Max Neural Network; Real Data;
Conference_Titel :
Control Conference (CCC), 2011 30th Chinese
Conference_Location :
Yantai
Print_ISBN :
978-1-4577-0677-6
Electronic_ISBN :
1934-1768