Title :
Investigation of automatic rule generation for hierarchical fuzzy systems
Author_Institution :
Bavarian Res. Center for Knowledge-Based Syst., Erlangen, Germany
Abstract :
The paper is based upon a method for the automatic generation of rules for hierarchical fuzzy associative memories (HIFAM). The effectiveness and termination of a general description of a training algorithm for HIFAMs is proved and experimental results on how HIFAMs evolve during training are given as well as a comparison with other machine learning techniques
Keywords :
content-addressable storage; function approximation; fuzzy neural nets; fuzzy systems; inference mechanisms; learning (artificial intelligence); pattern classification; automatic rule generation; hierarchical fuzzy associative memories; hierarchical fuzzy systems; machine learning techniques; training algorithm; Associative memory; Binary trees; Fuzzy sets; Fuzzy systems; Knowledge based systems; Machine learning; Machine learning algorithms; Quantization; Training data; Tree data structures;
Conference_Titel :
Fuzzy Systems Proceedings, 1998. IEEE World Congress on Computational Intelligence., The 1998 IEEE International Conference on
Conference_Location :
Anchorage, AK
Print_ISBN :
0-7803-4863-X
DOI :
10.1109/FUZZY.1998.686250
