Title :
FI-GEM networks for incomplete time-series prediction
Author :
Chiewchanwattana, Sirapat ; Lursinsap, Chidchanok
Author_Institution :
Dept. of Comput. Sci., Khon Kaen Univ., Thailand
fDate :
6/24/1905 12:00:00 AM
Abstract :
This paper considers the problem of incomplete time-series prediction by FI-GEM (fill-in-generalized ensemble method) networks, which has two steps. The first step is composed of several fill-in methods for preprocessing the missing value of time-series and the outcome are the complete time-series data. The second step is composed of the several individual multilayer perceptrons (MLP) whose their outputs are combined by the generalized ensemble method. There are five fill-in methods that are explored: cubic smoothing spline interpolation, and four imputation methods: EM (expectation maximization), regularized EM, average EM, average regularized EM. Mackey-Glass chaotic time-series and sunspots data are used for evaluating our approach. The experimental results show that the prediction accuracy of FI-GEM networks are much better than individual neural networks
Keywords :
forecasting theory; interpolation; multilayer perceptrons; splines (mathematics); time series; FI-GEM networks; MLP; Mackey-Glass chaotic time-series; average regularized EM; cubic smoothing spline interpolation; expectation maximization; fill-in-generalized ensemble method; imputation methods; incomplete time-series prediction; multilayer perceptrons; neural networks; sunspots data; Accuracy; Computer networks; Computer science; Intelligent networks; Interpolation; Mathematics; Multilayer perceptrons; Neural networks; Smoothing methods; Spline;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1007784