An improved ART1 neural network learning algorithm

Since inadequacy exists in the traditional ART1 neural network, an improved calculating method of similarity is proposed. This method considers the corresponding position value of the two vectors at the same time, thus avoiding different results of similarity because of different inputting sequence of the two vectors in the ART1 neural network. In order to solve the pattern-drifting problem of ART1 neural network, the principle of minority subordinate to majority is proposed to reduce the occurrence of this type of problem, improving the application effect of ART1 neural network.