Title :
Evolving fuzzy clusters
Author :
Fogel, David B. ; Simpson, Patrick K.
Author_Institution :
Orincon Corp., San Diego, CA, USA
Abstract :
An alternate training technique for the fuzzy min-max clustering neural network is introduced. The original fuzzy min-max clustering neural network utilized an algorithm similar to leader clustering and adaptive resonance theory to place hyperboxes in the pattern space. An alternative clustering technique is introduced, which utilizes evolutionary programming and information criteria to produce a set of hyperboxes
Keywords :
fuzzy set theory; learning (artificial intelligence); minimax techniques; neural nets; adaptive resonance theory; evolutionary programming; fuzzy min-max clustering neural network; hyperboxes; information criteria; leader clustering; learning; pattern space; Clustering algorithms; Function approximation; Fuzzy neural networks; Fuzzy sets; Genetic programming; Hypercubes; Neural networks; Resonance; Transfer functions;
Conference_Titel :
Neural Networks, 1993., IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0999-5
DOI :
10.1109/ICNN.1993.298835
