A Novel Nonlinear Combination Model Based on Support Vector Machine for Rainfall Prediction

Abstract

In this study, a novel modular-type Support Vector Machine (SVM) is presented to simulate rainfall prediction. First of all, a bagging sampling technique is used to generate different training sets. Secondly, different kernel function of SVM with different parameters, i.e., base models, are then trained to formulate different regression based on the different training sets. Thirdly, the Partial Least Square (PLS) technology is used to select choose the appropriate number of SVR combination members. Finally, a -SVM can be produced by learning from all base models. The technique will be implemented to forecast monthly rainfall in the Guangxi, China. Empirical results show that the prediction by using the SVM combination model is generally better than those obtained using other models presented in this study in terms of the same evaluation measurements. Our findings reveal that the nonlinear ensemble model proposed here can be used as an alternative forecasting tool for a Meteorological application in achieving greater forecasting accuracy and improving prediction quality further.

A Survey on Rainfall Prediction using Artificial Neural Network

“…Chadwick et al (2011) employed an artificial neural network approach to downscale GCM temperature and rainfall fields to regional model scale over Europe [43]. A novel modular-type Support Vector Machine (SVM) have presented by Lu and Wang (2011) to forecast monthly rainfall in the Guangxi, China. Ultimately, they have showed that the prediction by using the SVM combination model was generally better than those obtained using other models presented in terms of the same evaluation measurements.…”

“…Ultimately, they have showed that the prediction by using the SVM combination model was generally better than those obtained using other models presented in terms of the same evaluation measurements. The authors strongly believed that it could be used as an alternative forecasting tool for a Meteorological application in achieving greater forecasting accuracy and improving prediction quality further [44]. …”

A Review of Weather Forecasting Using Data Mining Techniques

“…It provides data to evaluate the efficiency water resource utilization, and provide reliable basis for local department to manage and plan. Kesheng and Lingzhi [10] presented a novel modular type support vector machine to simulate rainfall prediction. V-SVM regression model,which introduced a new parameter -V" which can control the number of support vectors and training errors without defining ϵ a prior.…”

Comparison of Stochastic and Machine Learning Methods for Multi-Step Ahead Forecasting of Hydrological Processes

“…The long list of studies using NN algorithms for hydrological forecasting tasks within a specific hydrological context includes Atiya et al (1999), Lambrakis et al (2000), Kişi (2007), Cheng et al (2008) and Yaseen et al (2016), while the reader can find relevant studies using SVM algorithms in Sivapragasam et al (2001), Shi andHan (2007), Kişi andCimen (2011) and Lu and Wang (2011); also some critical comments for such studies have been raised by Koutsoyiannis (2007). Furthermore, the literature contains a large number of studies proposing hybrid forecasting methods, e.g.…”

“…Another type of process attracting scientific and practical forecasting interest is precipitation (e.g. Sivapragasam et al (2001), Hong (2008), Pai and Hong (2007), Lu andWang (2011), Sivapragasam et al (2001) and Kişi and Cimen (2012)). Nevertheless, as emphasized in Zaini et al (2015) precipitation and streamflow are amongst the most difficult geophysical processes to forecast.…”