A Comparison of Gaussian Process and M5P for Prediction of Soil Permeability Coefficient

The permeability coefficient (k) of soil is one of the most important parameters affecting soil characteristics such as shear strength or settlement. Thus, determining soil permeability coefficient is very crucial; however, a field test for determining this parameter is difficult, time-consuming, an...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Binh Thai Pham, Hai-Bang Ly, Nadhir Al-Ansari, Lanh Si Ho
Formato:	article
Lenguaje:	EN
Publicado:	Hindawi Limited 2021
Materias:	Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/aa0336b21b2341ea99185a93038b23ee
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:aa0336b21b2341ea99185a93038b23ee
record_format	dspace
spelling	oai:doaj.org-article:aa0336b21b2341ea99185a93038b23ee2021-11-08T02:36:59ZA Comparison of Gaussian Process and M5P for Prediction of Soil Permeability Coefficient1875-919X10.1155/2021/3625289https://doaj.org/article/aa0336b21b2341ea99185a93038b23ee2021-01-01T00:00:00Zhttp://dx.doi.org/10.1155/2021/3625289https://doaj.org/toc/1875-919XThe permeability coefficient (k) of soil is one of the most important parameters affecting soil characteristics such as shear strength or settlement. Thus, determining soil permeability coefficient is very crucial; however, a field test for determining this parameter is difficult, time-consuming, and expensive. In this study, soft computing methods, namely, M5P and Gaussian process (GP), for estimating the permeability coefficient were constructed and compared. The results of this paper indicate that the two soft computing algorithms functioned well in predicting k. These two methods gave high accuracy of prediction capability. The determination coefficient of M5P (R2 = 0.766) was higher than that (R2 = 0.700) of GP. This implies that the M5P model is more reliable estimation than the GP model in predicting soils’ permeability coefficient (k). This proves that applying these machine learning techniques can provide an alternative for predicting basic soil parameters, including the permeability coefficient of soil.Binh Thai PhamHai-Bang LyNadhir Al-AnsariLanh Si HoHindawi LimitedarticleComputer softwareQA76.75-76.765ENScientific Programming, Vol 2021 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Computer software QA76.75-76.765
spellingShingle	Computer software QA76.75-76.765 Binh Thai Pham Hai-Bang Ly Nadhir Al-Ansari Lanh Si Ho A Comparison of Gaussian Process and M5P for Prediction of Soil Permeability Coefficient
description	The permeability coefficient (k) of soil is one of the most important parameters affecting soil characteristics such as shear strength or settlement. Thus, determining soil permeability coefficient is very crucial; however, a field test for determining this parameter is difficult, time-consuming, and expensive. In this study, soft computing methods, namely, M5P and Gaussian process (GP), for estimating the permeability coefficient were constructed and compared. The results of this paper indicate that the two soft computing algorithms functioned well in predicting k. These two methods gave high accuracy of prediction capability. The determination coefficient of M5P (R2 = 0.766) was higher than that (R2 = 0.700) of GP. This implies that the M5P model is more reliable estimation than the GP model in predicting soils’ permeability coefficient (k). This proves that applying these machine learning techniques can provide an alternative for predicting basic soil parameters, including the permeability coefficient of soil.
format	article
author	Binh Thai Pham Hai-Bang Ly Nadhir Al-Ansari Lanh Si Ho
author_facet	Binh Thai Pham Hai-Bang Ly Nadhir Al-Ansari Lanh Si Ho
author_sort	Binh Thai Pham
title	A Comparison of Gaussian Process and M5P for Prediction of Soil Permeability Coefficient
title_short	A Comparison of Gaussian Process and M5P for Prediction of Soil Permeability Coefficient
title_full	A Comparison of Gaussian Process and M5P for Prediction of Soil Permeability Coefficient
title_fullStr	A Comparison of Gaussian Process and M5P for Prediction of Soil Permeability Coefficient
title_full_unstemmed	A Comparison of Gaussian Process and M5P for Prediction of Soil Permeability Coefficient
title_sort	comparison of gaussian process and m5p for prediction of soil permeability coefficient
publisher	Hindawi Limited
publishDate	2021
url	https://doaj.org/article/aa0336b21b2341ea99185a93038b23ee
work_keys_str_mv	AT binhthaipham acomparisonofgaussianprocessandm5pforpredictionofsoilpermeabilitycoefficient AT haibangly acomparisonofgaussianprocessandm5pforpredictionofsoilpermeabilitycoefficient AT nadhiralansari acomparisonofgaussianprocessandm5pforpredictionofsoilpermeabilitycoefficient AT lanhsiho acomparisonofgaussianprocessandm5pforpredictionofsoilpermeabilitycoefficient AT binhthaipham comparisonofgaussianprocessandm5pforpredictionofsoilpermeabilitycoefficient AT haibangly comparisonofgaussianprocessandm5pforpredictionofsoilpermeabilitycoefficient AT nadhiralansari comparisonofgaussianprocessandm5pforpredictionofsoilpermeabilitycoefficient AT lanhsiho comparisonofgaussianprocessandm5pforpredictionofsoilpermeabilitycoefficient
_version_	1718443125944352768

A Comparison of Gaussian Process and M5P for Prediction of Soil Permeability Coefficient

Ejemplares similares