A study of hold-out and k-fold cross validation for accuracy of groundwater modeling in tidal lowland reclamation using extreme learning machine

Author

Nurhayati ; Hadihardaja, Iwan K. ; Soekarno, Indratmo ; Cahyono, M.

Author_Institution

Civil Eng. Dept., Univ. of Tanjungpura, Pontianak, Indonesia

fYear

2014

fDate

19-21 Aug. 2014

Firstpage

228

Lastpage

233

Abstract

The accuracy of prediction is required to conduct modeling the groundwater flow. This research represents the application of extreme learning machine (ELM) that can be used to model the groundwater flow in tidal lowland reclamation. The accuracy is measured using the hold-out and k-fold cross validation methods. The study will be implemented in the Delta Telang I Lowlands area, Banyuasin District, South Sumatra Province. The results of this groundwater flow modeling shows that the accuracy using the k-fold cross validation is better than the hold-out method. The values of accuracy level of the results of simulation-1 for training are: MSE = 0.000042091 and MAPE = 0.3165%. The values of accuracy level of the results of simulation-1 for testing are: MSE = 0.000093083 and MAPE = 0.4006%.

Keywords

geophysics computing; groundwater; learning (artificial intelligence); Banyuasin District; Delta Telang I Lowlands area; ELM; Indonesia; South Sumatra Province; extreme learning machine; groundwater modeling; hold-out cross validation; k-fold cross validation; tidal lowland reclamation; Accuracy; Data models; Predictive models; Simulation; Testing; Training; Training data; cross validation; extreme learning machine; groundwater; hold-out; tidal lowland;

fLanguage

English

Publisher

ieee

Conference_Titel

Technology, Informatics, Management, Engineering, and Environment (TIME-E), 2014 2nd International Conference on

Conference_Location

Bandung

Print_ISBN

978-1-4799-4806-2

Type

conf

DOI

10.1109/TIME-E.2014.7011623

Filename

7011623