The importance of sampling for the efficiency of artificial neural networks in digital soil modelling

In Portugal, soil mapping remains incomplete, and there are also significant problems with the existing cartography. Digital Soil Mapping uses advanced computerbased techniques such as Artificial Neural Networks (ANN) for mapping soil classes in a cheaper, more consistent and flexible way, using sur...

Full description

Bibliographic Details
Main Authors: Freire, Sérgio, Fonseca, Inês, Brasil, Ricardo, Rocha, Jorge, Tenedório, José António
Format: Book Part
Language:English
Published: Meubook 2023
Subjects:
Digital soil mapping
AutoMAPticS
IDRISI Taiga
Mondim de Basto
Vila Real
taiga
Online Access:http://hdl.handle.net/10451/57931
id ftunivlisboa:oai:repositorio.ul.pt:10451/57931
record_format openpolar
spelling ftunivlisboa:oai:repositorio.ul.pt:10451/57931 2023-06-18T03:43:17+02:00 The importance of sampling for the efficiency of artificial neural networks in digital soil modelling Freire, Sérgio Fonseca, Inês Brasil, Ricardo Rocha, Jorge Tenedório, José António 2023-06-04T17:43:17Z http://hdl.handle.net/10451/57931 eng eng Meubook Freire, S.; Fonseca, I.L.; Brasil, R.; Rocha, J. & Tenedório, J.A. (2012). The importance of sampling for the efficiency of artificial neural networks in digital soil modelling, In. D. Royé, J.A. Aldrey Vázquez, M. Pazoz Otón, M.J. Piñeira Mantiñán & M. Valcárcel Díaz (Eds), Actas de el XIII Coloquio Ibérico da Geografía, Respuestas de la Geografía Ibérica a la crisis actual, (pp. 867-876), Meubook, Santiago de Compostela, España. ISBN 978-84-940469-7-1. 978-84-940469-7-1 http://hdl.handle.net/10451/57931 openAccess Digital soil mapping AutoMAPticS IDRISI Taiga Mondim de Basto Vila Real bookPart 2023 ftunivlisboa 2023-06-07T00:11:04Z In Portugal, soil mapping remains incomplete, and there are also significant problems with the existing cartography. Digital Soil Mapping uses advanced computerbased techniques such as Artificial Neural Networks (ANN) for mapping soil classes in a cheaper, more consistent and flexible way, using surrogate landscape data. This work used five different training sets to evaluate the impact that sampling has on the predictive accuracy of ANNs. The testes were carried out in IDRISI Taiga for two catchments in northern Portugal, using an ANN method known as multi-layer perceptron. Results show that sampling design is very important for the accuracy of soil mapping with ANNs. info:eu-repo/semantics/publishedVersion Book Part taiga Universidade de Lisboa: repositório.UL
institution Open Polar
collection Universidade de Lisboa: repositório.UL
op_collection_id ftunivlisboa
language English
topic Digital soil mapping
AutoMAPticS
IDRISI Taiga
Mondim de Basto
Vila Real
spellingShingle Digital soil mapping
AutoMAPticS
IDRISI Taiga
Mondim de Basto
Vila Real
Freire, Sérgio
Fonseca, Inês
Brasil, Ricardo
Rocha, Jorge
Tenedório, José António
The importance of sampling for the efficiency of artificial neural networks in digital soil modelling
topic_facet Digital soil mapping
AutoMAPticS
IDRISI Taiga
Mondim de Basto
Vila Real
description In Portugal, soil mapping remains incomplete, and there are also significant problems with the existing cartography. Digital Soil Mapping uses advanced computerbased techniques such as Artificial Neural Networks (ANN) for mapping soil classes in a cheaper, more consistent and flexible way, using surrogate landscape data. This work used five different training sets to evaluate the impact that sampling has on the predictive accuracy of ANNs. The testes were carried out in IDRISI Taiga for two catchments in northern Portugal, using an ANN method known as multi-layer perceptron. Results show that sampling design is very important for the accuracy of soil mapping with ANNs. info:eu-repo/semantics/publishedVersion
format Book Part
author Freire, Sérgio
Fonseca, Inês
Brasil, Ricardo
Rocha, Jorge
Tenedório, José António
author_facet Freire, Sérgio
Fonseca, Inês
Brasil, Ricardo
Rocha, Jorge
Tenedório, José António
author_sort Freire, Sérgio
title The importance of sampling for the efficiency of artificial neural networks in digital soil modelling
title_short The importance of sampling for the efficiency of artificial neural networks in digital soil modelling
title_full The importance of sampling for the efficiency of artificial neural networks in digital soil modelling
title_fullStr The importance of sampling for the efficiency of artificial neural networks in digital soil modelling
title_full_unstemmed The importance of sampling for the efficiency of artificial neural networks in digital soil modelling
title_sort importance of sampling for the efficiency of artificial neural networks in digital soil modelling
publisher Meubook
publishDate 2023
url http://hdl.handle.net/10451/57931
genre taiga
genre_facet taiga
op_relation Freire, S.; Fonseca, I.L.; Brasil, R.; Rocha, J. & Tenedório, J.A. (2012). The importance of sampling for the efficiency of artificial neural networks in digital soil modelling, In. D. Royé, J.A. Aldrey Vázquez, M. Pazoz Otón, M.J. Piñeira Mantiñán & M. Valcárcel Díaz (Eds), Actas de el XIII Coloquio Ibérico da Geografía, Respuestas de la Geografía Ibérica a la crisis actual, (pp. 867-876), Meubook, Santiago de Compostela, España. ISBN 978-84-940469-7-1.
978-84-940469-7-1
http://hdl.handle.net/10451/57931
op_rights openAccess
_version_ 1769009615651995648

Similar Items