期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2019
卷号：19
期号：9
页码：82-90
出版社：International Journal of Computer Science and Network Security
摘要：Artificial neural network (ANN) is a powerful mathematical computational model, which is used in many important life applications such as function approximation, data regression, solving classification problem, pattern recognition and much other application. The objective of this paper is to use ANN for function approximation, to find the relationship from a given finite input-output data, that using in the many different application in this time. Different types of ANN will be created, trained and tested the obtained experimental results will be compared and discussed in order to select the best type of ANN, the best ANN architecture which will minimize the error between the targeted function value and the calculated one.
关键词：ANN; neuron; FFANN; cascade ANN; Elman ANN; ANN architecture; ANN parameters
