A data simulation system using YSINC polynomial higher order neural networks

Authors:
Ming Zhang
Affiliations:
Department of Physics, Computer Science and Engineering, Christopher Newport University, Newport News, Virginia
Venue:
MOAS'07 Proceedings of the 18th conference on Proceedings of the 18th IASTED International Conference: modelling and simulation
Year:
2007

Citing 3
Cited 0

A feedforward neural network with function shape autotuning

Neural Networks
Learning and approximation capabilities of adaptive spline activation function neutral networks

Neural Networks
Face recognition using artificial neural network group-based adaptive tolerance (GAT) trees

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper a new data simulation system, based on the y and sin(x)/x functions, called YSINC Polynomial Higher Order Neural Network (YSINCPHONN) has been developed. The YSINCPHONN model provides one more opportunity to find the optimal neural network model for simulation and prediction. This paper also proves that YSINCPHONN models can converge when simulating XOR data. This study also tests rainfall data. Based on the test results, YSINCPHONN model is 0.6562% better than Polynomial Higher Order Neural Network (PHONN) model and 0.62368% batter than trigonometric Polynomial Higher Order Neural Network (THONN) model in simulating rainfall. Moreover, foreign exchange rate simulation has been tested by using YSINCPHONN models. Test results show that YSINCPHONN models are about 0.7143% to 3.3243% better than PHONN, THONN, Sigmoid Polynomial Higher Order Neural Network (SPHONN), and SINC Polynomial Higher Order Neural Network (SINCHONN) models.