期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2016
卷号：7
期号：4
页码：1670-1673
出版社：TechScience Publications
摘要：The artificial neural network plays an importantrole in data classification. The increasing rate of data diversityneed efficient algorithm for the purpose of classification. Thecomplex architecture and weight adjustment process reducesthe performance of neural network classifier. In this paperproposed an optimum adjustment of weight and architectureusing particle swarm optimization. The particle swarmoptimization technique reduces the weight selection processand also optimized the hidden process layer of network. Theproposed algorithms implemented in mat lab software andtested well know dataset for the process of classification. Inthis paper used SOM neural network model for the weightadjustment factor and optimum architecture. The SOMneural network data mapped in two dimensional data spacefor the mapping of weight of input vector and processing datafor the process of clustering technique for the classification ofdata.
