Prediction of functional properties of nano $$\hbox {TiO}_2$$ TiO 2 coated cotton composites by artificial neural network

Abstract This paper represents the efficiency of machine learning tool, i.e., artificial neural network (ANN), for the prediction of functional properties of nano titanium dioxide coated cotton composites. A comparative analysis was performed between the predicted results of ANN, multiple linear reg...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Nesrine Amor, Muhammad Tayyab Noman, Michal Petru
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/dfb287ed822b417ab041078d8f3605ef
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:dfb287ed822b417ab041078d8f3605ef
record_format dspace
spelling oai:doaj.org-article:dfb287ed822b417ab041078d8f3605ef2021-12-02T14:59:29ZPrediction of functional properties of nano $$\hbox {TiO}_2$$ TiO 2 coated cotton composites by artificial neural network10.1038/s41598-021-91733-y2045-2322https://doaj.org/article/dfb287ed822b417ab041078d8f3605ef2021-06-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-91733-yhttps://doaj.org/toc/2045-2322Abstract This paper represents the efficiency of machine learning tool, i.e., artificial neural network (ANN), for the prediction of functional properties of nano titanium dioxide coated cotton composites. A comparative analysis was performed between the predicted results of ANN, multiple linear regression (MLR) and experimental results. ANN was applied to map out the complex input-output conditions to predict the optimal results. A backpropagation ANN model called a multilayer perceptron (MLP), trained with Bayesian regularization were used in this study. The amount of chemicals and reaction time were selected as input variables and the amount of titanium dioxide coated on cotton, self-cleaning efficiency, antimicrobial efficiency and ultraviolet protection factor were analysed as output results. The accuracy of the proposed algorithm was evaluated and compared with MLR results. The obtained results reveal that MLP provides efficient results that are statistically significant in the prediction of functional properties ( $$p<0.1 e^{-10} $$ p < 0.1 e - 10 ) compared to MLR. The correlation coefficient of MLP model ( $$>95\%$$ > 95 % ) indicates that there is a strong correlation between the measured and predicted functional properties with a trivial mean absolute error and root mean square errors values. MLP model is suitable for the functional properties and can be used for the investigation of other properties of nano coated fabrics.Nesrine AmorMuhammad Tayyab NomanMichal PetruNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Nesrine Amor
Muhammad Tayyab Noman
Michal Petru
Prediction of functional properties of nano $$\hbox {TiO}_2$$ TiO 2 coated cotton composites by artificial neural network
description Abstract This paper represents the efficiency of machine learning tool, i.e., artificial neural network (ANN), for the prediction of functional properties of nano titanium dioxide coated cotton composites. A comparative analysis was performed between the predicted results of ANN, multiple linear regression (MLR) and experimental results. ANN was applied to map out the complex input-output conditions to predict the optimal results. A backpropagation ANN model called a multilayer perceptron (MLP), trained with Bayesian regularization were used in this study. The amount of chemicals and reaction time were selected as input variables and the amount of titanium dioxide coated on cotton, self-cleaning efficiency, antimicrobial efficiency and ultraviolet protection factor were analysed as output results. The accuracy of the proposed algorithm was evaluated and compared with MLR results. The obtained results reveal that MLP provides efficient results that are statistically significant in the prediction of functional properties ( $$p<0.1 e^{-10} $$ p < 0.1 e - 10 ) compared to MLR. The correlation coefficient of MLP model ( $$>95\%$$ > 95 % ) indicates that there is a strong correlation between the measured and predicted functional properties with a trivial mean absolute error and root mean square errors values. MLP model is suitable for the functional properties and can be used for the investigation of other properties of nano coated fabrics.
format article
author Nesrine Amor
Muhammad Tayyab Noman
Michal Petru
author_facet Nesrine Amor
Muhammad Tayyab Noman
Michal Petru
author_sort Nesrine Amor
title Prediction of functional properties of nano $$\hbox {TiO}_2$$ TiO 2 coated cotton composites by artificial neural network
title_short Prediction of functional properties of nano $$\hbox {TiO}_2$$ TiO 2 coated cotton composites by artificial neural network
title_full Prediction of functional properties of nano $$\hbox {TiO}_2$$ TiO 2 coated cotton composites by artificial neural network
title_fullStr Prediction of functional properties of nano $$\hbox {TiO}_2$$ TiO 2 coated cotton composites by artificial neural network
title_full_unstemmed Prediction of functional properties of nano $$\hbox {TiO}_2$$ TiO 2 coated cotton composites by artificial neural network
title_sort prediction of functional properties of nano $$\hbox {tio}_2$$ tio 2 coated cotton composites by artificial neural network
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/dfb287ed822b417ab041078d8f3605ef
work_keys_str_mv AT nesrineamor predictionoffunctionalpropertiesofnanohboxtio2tio2coatedcottoncompositesbyartificialneuralnetwork
AT muhammadtayyabnoman predictionoffunctionalpropertiesofnanohboxtio2tio2coatedcottoncompositesbyartificialneuralnetwork
AT michalpetru predictionoffunctionalpropertiesofnanohboxtio2tio2coatedcottoncompositesbyartificialneuralnetwork
_version_ 1718389233242079232