ダウンロード数: 46

このアイテムのファイル:
ファイル 記述 サイズフォーマット 
27660400.2022.2109447.pdf6.28 MBAdobe PDF見る/開く
タイトル: Effects of data bias on machine-learning–based material discovery using experimental property data
著者: Kumagai, Masaya  kyouindb  KAKEN_id
Ando, Yuki
Tanaka, Atsumi
Tsuda, Koji
Katsura, Yukari
Kurosaki, Ken  kyouindb  KAKEN_id  orcid https://orcid.org/0000-0002-3015-3206 (unconfirmed)
著者名の別形: 熊谷, 将也
黒﨑, 健
キーワード: machine learning
material informatics
large-scale material data
data bias
発行日: Dec-2022
出版者: Taylor & Francis
誌名: Science and Technology of Advanced Materials: Methods
巻: 2
号: 1
開始ページ: 302
終了ページ: 309
抄録: Materials informatics (MI) research, which is the discovery of new materials through machine learning (ML) using large-scale material data, has attracted considerable attention in recent years. However, in general, the large-scale material data used in MI are biased owing to differences in the targeted material domains. Moreover, most studies on MI have not clearly demonstrated the influence of data bias on ML models. In this study, we clarify the influence of data bias on ML models by combining the concept of the applicability domain and clustering for large-scale experimental property data in the Starrydata2 material database previously developed by our group. The results show that data bias influences the error and reliability of the predictions made by the ML model. The predictions of the ML model within the applicability domain are highly reliable compared to those made outside the domain. This indicates that the material space that can be reliably discovered by the constructed ML model is limited. Nonetheless, we apply the ML model to a large dataset comprising various material classes and find that new materials similar to known materials can be proposed within a limited space. Thus, our findings demonstrate the importance of considering data bias when constructing and evaluating ML models in MI.
著作権等: © 2022 The Author(s). Published by National Institute for Materials Science in partnership with Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
URI: http://hdl.handle.net/2433/285724
DOI(出版社版): 10.1080/27660400.2022.2109447
出現コレクション:学術雑誌掲載論文等

アイテムの詳細レコードを表示する

Export to RefWorks


出力フォーマット 


このアイテムは次のライセンスが設定されています: クリエイティブ・コモンズ・ライセンス Creative Commons