Abstract—Training Artificial Neural Network (ANN) has
attracted many researchers for a long time. This paper
investigates the impact of training iterations of ANN using
Backpropagation Neural Network (BPNN) algorithm. The two
sets of adjustable parameters, i.e., the learning rate and
number of hidden nodes in the hidden layer are used to analyze
the impact of training iterations of ANN applications is used.
The applications that are used in this research are XOR
problem and Digit Recognition. The efficacy of the results
using BPNN algorithm is shown through an analysis of the
impact of training iterations and by presenting simulation
results from two different applications.
Index Terms—XOR, digit, training iterations, BPNN
The authors are with Iraq University, Islamabad, Pakistan (e-mail:
qamar.bhk@gmail.com, waqas_bangyal@hotmail.com,
jamil.ahmad@abasyn.edu.pk).
[PDF]
Cite:Qamar Abbas, Waqas Haider Bangyal, and Jamil Ahmad, "The Impact of Training Iterations on ANN Applications
Using BPNN Algorithm," International Journal of Future Computer and Communication vol. 2, no. 6, pp. 567-569, 2013.
