A supervised fuzzy hamming net trained by the genetic algorithm

A special fitness function designed for the fuzzy hamming net was proposed based on the evolutionary principle of the genetic algorithm, which could measure the performance of different combinations of the key parameters of that net. Then the genetic algorithm was applied to training the net to find the optimal combination of those parameters. By this means the blindness and randomicity of the manual adjustment, which was commonly used to train that net, could be eliminated, and the fuzzy hamming net could be automatically optimized by GA, moreover, the capacity and efficiency for classification and function mapping of the trained net were improved, too.