Title :
A study on grey RBF prediction model
Author :
Yuan, Jing-ling ; Zhong, Luo ; Li, Bing ; Jiang, Qiong
Author_Institution :
Sch. of Comput. Sci. & Technol., Wuhan Univ. of Technol., China
Abstract :
A new prediction model of small sample for time-displacement data based on RBF and GM(0, N), called grey RBF (GRBF) prediction model, was proposed in this paper. RBF network has good ability in approaching nonlinear function, and its convergent speed is rapid, but its precision and stability strongly depended on sample capability. While GM(0, N) can make an accurate precision of small sample data, unfortunately it is a linear static model. The new model here this paper proposed has combined the two above-mentioned models, which can give full play to their advantages and work more effectively. The new prediction model was used to predict other key points´ displacement of the same time in terms of the limited displacement of certain time. The result of experiment shows that this method is easy, convenient, accurate and it has a good practicality as well.
Keywords :
convergence; data analysis; grey systems; radial basis function networks; convergence; grey RBF prediction model; linear static model; nonlinear function; radial basis function neural network; small sample time-displacement data; Computer science; Data engineering; Displacement measurement; Electronic mail; Equations; Machine learning; Neural networks; Predictive models; Radial basis function networks; Stability; GM (0, N) model; Grey RBF (GRBF) Prediction Model; RBF neural network; Small sample Time-Displacement Data;
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
DOI :
10.1109/ICMLC.2005.1527662
