An Algorithm for Fuzzy Pattern Recognition Based on Neural Networks

摘要

According to principles of fuzzy mathematics and neural networks, a new model on neural networks, by which fuzzy patterns can be better recognized, is presented in this paper. This model combines the thoughts of neural networks and maximum membership function. Thus the insufficiency in semantic expressions of neural networks can be compensated for. In the meantime, more objective effect can be obtained than that by fuzzy pattern recognition method in fuzzy mathematics. Experimental results show that the method is valid in practical applications.