Computer Science and Information Engineering, World Congress on (2009)
Los Angeles, California USA
Mar. 31, 2009 to Apr. 2, 2009
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CSIE.2009.887
Neural networks with switching units were originally designed for classification tasks. However it was also shown that simple neural networks with switching units are capable to predict seasonal time series with results comparable to common stochastic methods. This paper presents enhanced model of neural network with switching units with aim to improve the forecasting performance of non-stationary time series. The presented model of neural network network is build of neurons with feedback and continuous activation function and it has a two level topology. The paper further describes the application of genetic algorithms to the optimization of the first level of topology. Finally, the performance of the proposed model was tested on the time series of currency in circulation and two artificial seasonal stochastic processes. Experimental results confirm that the new model outperforms the basic one as well as common stochastic methods.
switching unit, neural network, seasonal time series, currency in circulation
F. Hakl, R. Kalous and M. Hlavácek, "Neural Network with Cooperative Switching Units with Application to Time Series Forecasting," 2009 WRI World Congress on Computer Science and Information Engineering, CSIE(CSIE), Los Angeles, CA, 2009, pp. 676-682.