Improved validation framework and R-package for artificial neural network models

Validation is a critical component of any modelling process. In artificial neural network (ANN) modelling, validation generally consists of the assessment of model predictive performance on an independent validation set (predictive validity). However, this ignores other aspects of model validation c...

Full description

Bibliographic Details
Main Authors: Humphrey, Greer B., Maier, Holger R., Wu, Wenyan, Mount, Nick J., Dandy, Graeme C., Abrahart, R.J., Dawson, C.W.
Format: Article
Published: Elsevier 2017
Subjects:
Online Access:https://eprints.nottingham.ac.uk/40959/
_version_ 1848796172288786432
author Humphrey, Greer B.
Maier, Holger R.
Wu, Wenyan
Mount, Nick J.
Dandy, Graeme C.
Abrahart, R.J.
Dawson, C.W.
author_facet Humphrey, Greer B.
Maier, Holger R.
Wu, Wenyan
Mount, Nick J.
Dandy, Graeme C.
Abrahart, R.J.
Dawson, C.W.
author_sort Humphrey, Greer B.
building Nottingham Research Data Repository
collection Online Access
description Validation is a critical component of any modelling process. In artificial neural network (ANN) modelling, validation generally consists of the assessment of model predictive performance on an independent validation set (predictive validity). However, this ignores other aspects of model validation considered to be good practice in other areas of environmental modelling, such as residual analysis (replicative validity) and checking the plausibility of the model in relation to a priori system understanding (structural validity). In order to address this shortcoming, a validation framework for ANNs is introduced in this paper that covers all of the above aspects of validation. In addition, the validann R-package is introduced that enables these validation methods to be implemented in a user-friendly and consistent fashion. The benefits of the framework and R-package are demonstrated for two environmental modelling case studies, highlighting the importance of considering replicative and structural validity in addition to predictive validity.
first_indexed 2025-11-14T19:43:45Z
format Article
id nottingham-40959
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:43:45Z
publishDate 2017
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling nottingham-409592020-05-04T18:53:14Z https://eprints.nottingham.ac.uk/40959/ Improved validation framework and R-package for artificial neural network models Humphrey, Greer B. Maier, Holger R. Wu, Wenyan Mount, Nick J. Dandy, Graeme C. Abrahart, R.J. Dawson, C.W. Validation is a critical component of any modelling process. In artificial neural network (ANN) modelling, validation generally consists of the assessment of model predictive performance on an independent validation set (predictive validity). However, this ignores other aspects of model validation considered to be good practice in other areas of environmental modelling, such as residual analysis (replicative validity) and checking the plausibility of the model in relation to a priori system understanding (structural validity). In order to address this shortcoming, a validation framework for ANNs is introduced in this paper that covers all of the above aspects of validation. In addition, the validann R-package is introduced that enables these validation methods to be implemented in a user-friendly and consistent fashion. The benefits of the framework and R-package are demonstrated for two environmental modelling case studies, highlighting the importance of considering replicative and structural validity in addition to predictive validity. Elsevier 2017-06-30 Article PeerReviewed Humphrey, Greer B., Maier, Holger R., Wu, Wenyan, Mount, Nick J., Dandy, Graeme C., Abrahart, R.J. and Dawson, C.W. (2017) Improved validation framework and R-package for artificial neural network models. Environmental Modelling and Software, 92 . pp. 82-106. ISSN 1364-8152 Artificial neural networks; Multi-layer perceptron; R-package; Structural validation; Replicative validation; Predictive validation http://www.sciencedirect.com/science/article/pii/S136481521630963X doi:10.1016/j.envsoft.2017.01.023 doi:10.1016/j.envsoft.2017.01.023
spellingShingle Artificial neural networks; Multi-layer perceptron; R-package; Structural validation; Replicative validation; Predictive validation
Humphrey, Greer B.
Maier, Holger R.
Wu, Wenyan
Mount, Nick J.
Dandy, Graeme C.
Abrahart, R.J.
Dawson, C.W.
Improved validation framework and R-package for artificial neural network models
title Improved validation framework and R-package for artificial neural network models
title_full Improved validation framework and R-package for artificial neural network models
title_fullStr Improved validation framework and R-package for artificial neural network models
title_full_unstemmed Improved validation framework and R-package for artificial neural network models
title_short Improved validation framework and R-package for artificial neural network models
title_sort improved validation framework and r-package for artificial neural network models
topic Artificial neural networks; Multi-layer perceptron; R-package; Structural validation; Replicative validation; Predictive validation
url https://eprints.nottingham.ac.uk/40959/
https://eprints.nottingham.ac.uk/40959/
https://eprints.nottingham.ac.uk/40959/