Regularized versus non-regularized neural network model for prediction of saturated soil-water content on weathered granite soil formation

Muhammad Mukhlisin, Ahmed El-Shafie, Mohd. Raihan Taha

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Modeling unsaturated water flow in soil requires knowledge of the hydraulic properties of soil. However, correlation between soil hydraulic properties such as the relationship between saturated soil-water content θ s and saturated soil hydraulic conductivity k s as function of soil depth is in stochastic pattern. On the other hand, soil-water profile process is also believed to be highly non-linear, time varying, spatially distributed, and not easily described by simple models. In this study, the potential of implementing artificial neural network (ANN) model was proposed and investigated to map the soil-water profile in terms of k s and θ s with respect to the soil depth d. A regularized neural network (NN) model is developed to overcome the drawbacks of conventional prediction techniques. The use of regularized NN advantaged avoid over-fitting of training data, which was observed as a limitation of classical ANN models. Site experimental data sets on the hydraulic properties of weathered granite soils were collected. These data sets include the observed values of saturated and unsaturated hydraulic conductivities, saturated water contents, and retention curves. The proposed ANN model was examined utilizing 49 records of data collected from field experiments. The results showed that the regularized ANN model has the ability to detect and extract the stochastic behavior of saturated soil-water content with relatively high accuracy.

Original languageEnglish
Pages (from-to)543-553
Number of pages11
JournalNeural Computing and Applications
Volume21
Issue number3
DOIs
Publication statusPublished - Apr 2012

Fingerprint

Granite
Water content
Neural networks
Soils
Hydraulic conductivity
Hydraulics
Water

Keywords

  • Over-fitting
  • Regularized neural network
  • Saturated soil-water content
  • Soil formation

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software

Cite this

Regularized versus non-regularized neural network model for prediction of saturated soil-water content on weathered granite soil formation. / Mukhlisin, Muhammad; El-Shafie, Ahmed; Taha, Mohd. Raihan.

In: Neural Computing and Applications, Vol. 21, No. 3, 04.2012, p. 543-553.

Research output: Contribution to journalArticle

@article{2ad948ddc5464f02b62666b51c5c295b,
title = "Regularized versus non-regularized neural network model for prediction of saturated soil-water content on weathered granite soil formation",
abstract = "Modeling unsaturated water flow in soil requires knowledge of the hydraulic properties of soil. However, correlation between soil hydraulic properties such as the relationship between saturated soil-water content θ s and saturated soil hydraulic conductivity k s as function of soil depth is in stochastic pattern. On the other hand, soil-water profile process is also believed to be highly non-linear, time varying, spatially distributed, and not easily described by simple models. In this study, the potential of implementing artificial neural network (ANN) model was proposed and investigated to map the soil-water profile in terms of k s and θ s with respect to the soil depth d. A regularized neural network (NN) model is developed to overcome the drawbacks of conventional prediction techniques. The use of regularized NN advantaged avoid over-fitting of training data, which was observed as a limitation of classical ANN models. Site experimental data sets on the hydraulic properties of weathered granite soils were collected. These data sets include the observed values of saturated and unsaturated hydraulic conductivities, saturated water contents, and retention curves. The proposed ANN model was examined utilizing 49 records of data collected from field experiments. The results showed that the regularized ANN model has the ability to detect and extract the stochastic behavior of saturated soil-water content with relatively high accuracy.",
keywords = "Over-fitting, Regularized neural network, Saturated soil-water content, Soil formation",
author = "Muhammad Mukhlisin and Ahmed El-Shafie and Taha, {Mohd. Raihan}",
year = "2012",
month = "4",
doi = "10.1007/s00521-011-0545-2",
language = "English",
volume = "21",
pages = "543--553",
journal = "Neural Computing and Applications",
issn = "0941-0643",
publisher = "Springer London",
number = "3",

}

TY - JOUR

T1 - Regularized versus non-regularized neural network model for prediction of saturated soil-water content on weathered granite soil formation

AU - Mukhlisin, Muhammad

AU - El-Shafie, Ahmed

AU - Taha, Mohd. Raihan

PY - 2012/4

Y1 - 2012/4

N2 - Modeling unsaturated water flow in soil requires knowledge of the hydraulic properties of soil. However, correlation between soil hydraulic properties such as the relationship between saturated soil-water content θ s and saturated soil hydraulic conductivity k s as function of soil depth is in stochastic pattern. On the other hand, soil-water profile process is also believed to be highly non-linear, time varying, spatially distributed, and not easily described by simple models. In this study, the potential of implementing artificial neural network (ANN) model was proposed and investigated to map the soil-water profile in terms of k s and θ s with respect to the soil depth d. A regularized neural network (NN) model is developed to overcome the drawbacks of conventional prediction techniques. The use of regularized NN advantaged avoid over-fitting of training data, which was observed as a limitation of classical ANN models. Site experimental data sets on the hydraulic properties of weathered granite soils were collected. These data sets include the observed values of saturated and unsaturated hydraulic conductivities, saturated water contents, and retention curves. The proposed ANN model was examined utilizing 49 records of data collected from field experiments. The results showed that the regularized ANN model has the ability to detect and extract the stochastic behavior of saturated soil-water content with relatively high accuracy.

AB - Modeling unsaturated water flow in soil requires knowledge of the hydraulic properties of soil. However, correlation between soil hydraulic properties such as the relationship between saturated soil-water content θ s and saturated soil hydraulic conductivity k s as function of soil depth is in stochastic pattern. On the other hand, soil-water profile process is also believed to be highly non-linear, time varying, spatially distributed, and not easily described by simple models. In this study, the potential of implementing artificial neural network (ANN) model was proposed and investigated to map the soil-water profile in terms of k s and θ s with respect to the soil depth d. A regularized neural network (NN) model is developed to overcome the drawbacks of conventional prediction techniques. The use of regularized NN advantaged avoid over-fitting of training data, which was observed as a limitation of classical ANN models. Site experimental data sets on the hydraulic properties of weathered granite soils were collected. These data sets include the observed values of saturated and unsaturated hydraulic conductivities, saturated water contents, and retention curves. The proposed ANN model was examined utilizing 49 records of data collected from field experiments. The results showed that the regularized ANN model has the ability to detect and extract the stochastic behavior of saturated soil-water content with relatively high accuracy.

KW - Over-fitting

KW - Regularized neural network

KW - Saturated soil-water content

KW - Soil formation

UR - http://www.scopus.com/inward/record.url?scp=84858153709&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84858153709&partnerID=8YFLogxK

U2 - 10.1007/s00521-011-0545-2

DO - 10.1007/s00521-011-0545-2

M3 - Article

AN - SCOPUS:84858153709

VL - 21

SP - 543

EP - 553

JO - Neural Computing and Applications

JF - Neural Computing and Applications

SN - 0941-0643

IS - 3

ER -