Forecasting time series water levels
on Mekong river
using machine learning models
Faculty of Computer Scienceand Engineering,
Quynh Nguyen Huu
Information Technology Faculty,
Electric Power University
Abstract—Forecasting water levels on Mekong river is an important problem needed to be studied for ﬂood warning. In this paper, we investigate the application to forecasting of daily water levels at Thakhek station on Mekong river using machine learning models such as LASSO, Random Forests and Support Vector Regression (SVR). Experimental results showed that SVR was able to achieve feasible results, the mean absolute error of SVR is while the acceptable error of a ﬂood forecast model required by the Mekong River Commission is between 0.5m and 0.75m.
Please find the full article enclosed!
Veuillez trouver ci-joint ce plein article!