Euclidean ART Neural Networks
Riyadh Kenaya, Ka C. Cheok · 2008
Abstract—Unsupervised Neural networks are known for their ability to cluster input vectors (patterns) into categories (neurons) based on a neighborhood of similarity between two or more patterns and how big the radius of similarity to bet set by the network user. Fuzzy ART neural networks are examples of such systems where normalized input patterns are clustered into categories. It has been proven however that fuzzy ART networks show a disorder in their clustering performance especially when they are trained to learn noisy patterns. While Fuzzy ART networks employ the fuzzy AND neighborhood to