Groundwater Level Forecasting Based on Support Vector Machine

Abstract:

In this paper, we apply support vector regression (SVR) for groundwater level forecasting and compare its results to other prediction methods using real groundwater level data. Since support vector machines have greater generalization ability and guarantee global minima for given training data, it is believed that support vector regression will perform well for time series analysis. Compared to other predictors, our results show that the SVR predictor can reduce significantly both relative mean errors and root mean squared errors of predicted groundwater level. We demonstrate the feasibility of applying SVR in groundwater level prediction and prove that SVR is applicable and performs well for groundwater data analysis.

• Periodical: Applied Mechanics and Materials (Volumes 44-47)
• Pages: 1365-1369
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.44-47.1365
• Online since: December 2010
• Authors: Wei Guo Zhao, Huan Wang, Zi Jun Wang
• Keywords: Artificial Neural Network (ANN), Groundwater Level Forecast, Support Vector Machine (SVM), Support Vector Regression (SVR)

References

[1] Ioannis, N., Daliakopoulos, Coulibaly, P., Ioannis, K. and Tsanis.J. Hydrol. 2005, 309, 229-240.

[2] Hornik, K., Stinchcombe, M. and White, H., Neural Networks, 1989, 2, 359-366.

[3] V.N. Vapnik. The Nature of Statistical Learning Theory. Springer, New York, (1995).

[4] K.R. Muller, A. Smola, G. Ratch, B. Scholkopf, J. Kohlmorgen, and V Vapnik, Image Processing Services Research Lab, AT&T Labs.

[5] C.H. Wu IEEE Transactions on Intelligent Transportation Systems 5, (2004)276-281.

[6] Liu G, Ding J. Journal of Xian Geology College, 1997; 2: 45-50.

[7] Wang W S, Ding J. Nature and Science, 2003, 1(1): 67-71.