Analysis of Prediction Mechanisms and Feature Importance of Martensite Start Temperature of Alloy Steel via Explainable Artificial Intelligence

Junhyub Jeon; Namhyuk Seo; Jae Gil Jung; Seung Bae Son; Seok Jae Lee

doi:10.2320/matertrans.MT-M2023019

Analysis of Prediction Mechanisms and Feature Importance of Martensite Start Temperature of Alloy Steel via Explainable Artificial Intelligence

Junhyub Jeon
, Namhyuk Seo
, Jae Gil Jung
, Seung Bae Son
, Seok Jae Lee^*

^*Corresponding author for this work

Metallurgical System Engineering

Jeonbuk National University

Research output: Contribution to journal › Journal article › peer-review

4 Scopus citations

Abstract

This study proposes a machine learning model to predict the martensite start temperature (M_s) of alloy steels. We collected 219 usable data from the literature, and adjusted the hyperparameters to propose an accurate machine learning model. Artificial neural networks (ANN) exhibited the best performance compared with existing empirical equation. The prediction mechanisms and feature importance of the ANN with regards to the whole system were discussed via the Shapley additive explanation (SHAP).

Original language	English
Pages (from-to)	2196-2201
Number of pages	6
Journal	Materials Transactions
Volume	64
Issue number	9
DOIs	https://doi.org/10.2320/matertrans.MT-M2023019
State	Published - 2023

Keywords

alloy steels
explainable artificial intelligence
machine learning
martensite start temperature
prediction mechanism

Quacquarelli Symonds(QS) Subject Topics

Materials Science
Engineering - Mechanical
Physics & Astronomy

Access to Document

10.2320/matertrans.MT-M2023019

Cite this

@article{9b1235d1e5b84b5fbfd5daa5c891972e,

title = "Analysis of Prediction Mechanisms and Feature Importance of Martensite Start Temperature of Alloy Steel via Explainable Artificial Intelligence",

abstract = "This study proposes a machine learning model to predict the martensite start temperature (Ms) of alloy steels. We collected 219 usable data from the literature, and adjusted the hyperparameters to propose an accurate machine learning model. Artificial neural networks (ANN) exhibited the best performance compared with existing empirical equation. The prediction mechanisms and feature importance of the ANN with regards to the whole system were discussed via the Shapley additive explanation (SHAP).",

keywords = "alloy steels, explainable artificial intelligence, machine learning, martensite start temperature, prediction mechanism",

author = "Junhyub Jeon and Namhyuk Seo and Jung, \{Jae Gil\} and Son, \{Seung Bae\} and Lee, \{Seok Jae\}",

note = "Publisher Copyright: {\textcopyright}2023 The Japan Institute of Metals and Materials.",

year = "2023",

doi = "10.2320/matertrans.MT-M2023019",

language = "English",

volume = "64",

pages = "2196--2201",

journal = "Materials Transactions",

issn = "1345-9678",

number = "9",

}

TY - JOUR

T1 - Analysis of Prediction Mechanisms and Feature Importance of Martensite Start Temperature of Alloy Steel via Explainable Artificial Intelligence

AU - Jeon, Junhyub

AU - Seo, Namhyuk

AU - Jung, Jae Gil

AU - Son, Seung Bae

AU - Lee, Seok Jae

PY - 2023

Y1 - 2023

N2 - This study proposes a machine learning model to predict the martensite start temperature (Ms) of alloy steels. We collected 219 usable data from the literature, and adjusted the hyperparameters to propose an accurate machine learning model. Artificial neural networks (ANN) exhibited the best performance compared with existing empirical equation. The prediction mechanisms and feature importance of the ANN with regards to the whole system were discussed via the Shapley additive explanation (SHAP).

AB - This study proposes a machine learning model to predict the martensite start temperature (Ms) of alloy steels. We collected 219 usable data from the literature, and adjusted the hyperparameters to propose an accurate machine learning model. Artificial neural networks (ANN) exhibited the best performance compared with existing empirical equation. The prediction mechanisms and feature importance of the ANN with regards to the whole system were discussed via the Shapley additive explanation (SHAP).

KW - alloy steels

KW - explainable artificial intelligence

KW - machine learning

KW - martensite start temperature

KW - prediction mechanism

UR - https://www.scopus.com/pages/publications/85174406146

U2 - 10.2320/matertrans.MT-M2023019

DO - 10.2320/matertrans.MT-M2023019

M3 - Journal article

AN - SCOPUS:85174406146

SN - 1345-9678

VL - 64

SP - 2196

EP - 2201

JO - Materials Transactions

JF - Materials Transactions

IS - 9

ER -

Analysis of Prediction Mechanisms and Feature Importance of Martensite Start Temperature of Alloy Steel via Explainable Artificial Intelligence

Abstract

Keywords

Quacquarelli Symonds(QS) Subject Topics

Access to Document

Other files and links

Fingerprint

Cite this