作者: Chin-Der Wann , Stelios C. A. Thomopoulos
DOI: 10.1007/3-540-56798-4_166
关键词:
摘要: A benchmark study of self-organizing neural network models is conducted. The comparison advantages and disadvantages unsupervised learning artificial networks are discussed. discussed in this paper include adaptive resonance theory (ART2), DIGNET, feature map, vector quantization (LVQ). For the applications on data clustering pattern recognition problems with additive gaussian noise, we compare performance systems, ART2 DIGNET. Results computer simulation show that both DIGNET achieve good clustering, but faster process has better results overall performance.