Explainable machine learning and feature engineering applied to nanoindentation data

The work aims to challenge the hegemony in the literature of clustering nanoindentation data solely relying on elastic modulus and hardness as features, thereby discarding information provided by the full load–displacement curve. Features based on dimensional analysis initially aimed to solve the in...

Full description

Saved in:
Bibliographic Details
Main Authors: C.O.W. Trost, S. Žák, S. Schaffer, L. Walch, J. Zitz, T. Klünsner, H. Leitner, L. Exl, M.J. Cordill
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:Materials & Design
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S026412752500317X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850240700925345792
author C.O.W. Trost
S. Žák
S. Schaffer
L. Walch
J. Zitz
T. Klünsner
H. Leitner
L. Exl
M.J. Cordill
author_facet C.O.W. Trost
S. Žák
S. Schaffer
L. Walch
J. Zitz
T. Klünsner
H. Leitner
L. Exl
M.J. Cordill
author_sort C.O.W. Trost
collection DOAJ
description The work aims to challenge the hegemony in the literature of clustering nanoindentation data solely relying on elastic modulus and hardness as features, thereby discarding information provided by the full load–displacement curve. Features based on dimensional analysis initially aimed to solve the inverse nanoindentation problem were adopted to describe the load–displacement curves. More than 3000 indents in high-speed steels were labelled via imaging after indenting. The resulting dataset was used to train and benchmark supervised (classification) and unsupervised (clustering) machine learning models, showing that feature engineering was more impactful than model selection and hyperparameter tuning, increasing the prediction quality in all studied models. The best classifier’s predictions were explained via a game theory-based approach, allowing insights into the model’s decision-making process and connecting the fields of materials property clustering and materials mechanics.
format Article
id doaj-art-dea9d2e3aafa4366b2a0944c9c3c400c
institution OA Journals
issn 0264-1275
language English
publishDate 2025-05-01
publisher Elsevier
record_format Article
series Materials & Design
spelling doaj-art-dea9d2e3aafa4366b2a0944c9c3c400c2025-08-20T02:00:47ZengElsevierMaterials & Design0264-12752025-05-0125311389710.1016/j.matdes.2025.113897Explainable machine learning and feature engineering applied to nanoindentation dataC.O.W. Trost0S. Žák1S. Schaffer2L. Walch3J. Zitz4T. Klünsner5H. Leitner6L. Exl7M.J. Cordill8Erich Schmid Institute of Materials Science, Austrian Academy of Sciences, Jahnstrasse 12, 8700 Leoben, Austria; Corresponding author.Erich Schmid Institute of Materials Science, Austrian Academy of Sciences, Jahnstrasse 12, 8700 Leoben, AustriaWolfgang Pauli Institute c/o Faculty of Mathematics, University of Vienna, Oskar-Morgenstern-Platz 1, 1090 Vienna, Austria; University of Vienna Research Platform MMM Mathematics - Magnetism - Materials, University of Vienna, Oskar-Morgenstern-Platz 1, 1090 Vienna, AustriaMaterials Center, Leoben Forschung GmbH, Leoben, AustriaMaterials Center, Leoben Forschung GmbH, Leoben, AustriaMaterials Center, Leoben Forschung GmbH, Leoben, AustriaVoestalpine BÖHLER Edelstahl GmbH & Co KG, Mariazeller-Straße 25, A-8605 Kapfenberg, AustriaWolfgang Pauli Institute c/o Faculty of Mathematics, University of Vienna, Oskar-Morgenstern-Platz 1, 1090 Vienna, Austria; University of Vienna Research Platform MMM Mathematics - Magnetism - Materials, University of Vienna, Oskar-Morgenstern-Platz 1, 1090 Vienna, AustriaErich Schmid Institute of Materials Science, Austrian Academy of Sciences, Jahnstrasse 12, 8700 Leoben, Austria; Department of Materials Science, Montanuniversität Leoben, Jahnstrasse 12, Leoben 8700, AustriaThe work aims to challenge the hegemony in the literature of clustering nanoindentation data solely relying on elastic modulus and hardness as features, thereby discarding information provided by the full load–displacement curve. Features based on dimensional analysis initially aimed to solve the inverse nanoindentation problem were adopted to describe the load–displacement curves. More than 3000 indents in high-speed steels were labelled via imaging after indenting. The resulting dataset was used to train and benchmark supervised (classification) and unsupervised (clustering) machine learning models, showing that feature engineering was more impactful than model selection and hyperparameter tuning, increasing the prediction quality in all studied models. The best classifier’s predictions were explained via a game theory-based approach, allowing insights into the model’s decision-making process and connecting the fields of materials property clustering and materials mechanics.http://www.sciencedirect.com/science/article/pii/S026412752500317XNanoindentationExplainable machine learningHigh-speed steelComposite material
spellingShingle C.O.W. Trost
S. Žák
S. Schaffer
L. Walch
J. Zitz
T. Klünsner
H. Leitner
L. Exl
M.J. Cordill
Explainable machine learning and feature engineering applied to nanoindentation data
Materials & Design
Nanoindentation
Explainable machine learning
High-speed steel
Composite material
title Explainable machine learning and feature engineering applied to nanoindentation data
title_full Explainable machine learning and feature engineering applied to nanoindentation data
title_fullStr Explainable machine learning and feature engineering applied to nanoindentation data
title_full_unstemmed Explainable machine learning and feature engineering applied to nanoindentation data
title_short Explainable machine learning and feature engineering applied to nanoindentation data
title_sort explainable machine learning and feature engineering applied to nanoindentation data
topic Nanoindentation
Explainable machine learning
High-speed steel
Composite material
url http://www.sciencedirect.com/science/article/pii/S026412752500317X
work_keys_str_mv AT cowtrost explainablemachinelearningandfeatureengineeringappliedtonanoindentationdata
AT szak explainablemachinelearningandfeatureengineeringappliedtonanoindentationdata
AT sschaffer explainablemachinelearningandfeatureengineeringappliedtonanoindentationdata
AT lwalch explainablemachinelearningandfeatureengineeringappliedtonanoindentationdata
AT jzitz explainablemachinelearningandfeatureengineeringappliedtonanoindentationdata
AT tklunsner explainablemachinelearningandfeatureengineeringappliedtonanoindentationdata
AT hleitner explainablemachinelearningandfeatureengineeringappliedtonanoindentationdata
AT lexl explainablemachinelearningandfeatureengineeringappliedtonanoindentationdata
AT mjcordill explainablemachinelearningandfeatureengineeringappliedtonanoindentationdata