Supervised Fuzzy Neural Networks with Considering the Locations of Prototypes for Learning Rules
Yong Soo Kim · 한국지능시스템학회 국제학술대회 발표논문집 · 2007
In this paper, supervised fuzzy neural networks are presented. These supervised fuzzy neural networks use modified LVQ(Learning Vector Quantization) learning rules. The learning rates are fuzzified based on considering the location of the input vector compared to the locations of prototypes. These new fuzzy learning rules are integrated into IAFC(Integrated Adaptive Fuzzy Clustering) neural network. The performances of these supervised fuzzy neural networks are compared with those of other supervised neural networks.