Options
Predicting the inhibition efficiencies of magnesium dissolution modulators using sparse machine learning models
Citation Link: https://doi.org/10.15480/882.4040
Publikationstyp
Journal Article
Date Issued
2021-12
Sprache
English
Author(s)
TORE-DOI
Journal
Volume
7
Issue
1
Article Number
193
Citation
npj Computational Materials 7 (1): 193 (2021-12)
Publisher DOI
Scopus ID
Publisher
Nature Publ. Group
The degradation behaviour of magnesium and its alloys can be tuned by small organic molecules. However, an automatic identification of effective organic additives within the vast chemical space of potential compounds needs sophisticated tools. Herein, we propose two systematic approaches of sparse feature selection for identifying molecular descriptors that are most relevant for the corrosion inhibition efficiency of chemical compounds. One is based on the classical statistical tool of analysis of variance, the other one based on random forests. We demonstrate how both can—when combined with deep neural networks—help to predict the corrosion inhibition efficiencies of chemical compounds for the magnesium alloy ZE41. In particular, we demonstrate that this framework outperforms predictions relying on a random selection of molecular descriptors. Finally, we point out how autoencoders could be used in the future to enable even more accurate automated predictions of corrosion inhibition efficiencies.
Subjects
MLE@TUHH
DDC Class
004: Informatik
570: Biowissenschaften, Biologie
Funding Organisations
More Funding Information
Funding by the Helmholtz Association is gratefully acknowledged. T.W. and C.F. gratefully acknowledge funding by the Deutscher Akademischer Austauschdienst (DAAD, German Academic Exchange Service) via Projektnummer 57511455. R.M.
gratefully acknowledges funding by the Deutsche Forschungsgemeinschaft (D.F.G., German Research Foundation) via Projektnummer 192346071-SFB 986 and Projektnummer 390794421-GRK 2462.
gratefully acknowledges funding by the Deutsche Forschungsgemeinschaft (D.F.G., German Research Foundation) via Projektnummer 192346071-SFB 986 and Projektnummer 390794421-GRK 2462.
Publication version
publishedVersion
Loading...
Name
s41524-021-00658-7.pdf
Size
1.79 MB
Format
Adobe PDF