Data-Driven Approach Considering Imbalance in Data Sets and Experimental Conditions for Exploration of Photocatalysts
ACS Omega · 22.04.2025 · 10 (15) · 14626-14639
DOI: 10.1021/acsomega.4c06997 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACS Omega, online erschienen am 10.04.2025. Verfasst von Wataru Takahara und 7 weiteren Autor:innen. Beteiligte Einrichtung: Division of Materials Science, Nara Institute of Science and Technology, Ikoma-shi, Nara-ken 630-0192, Japan. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Ministry of Education, Culture, Sports, Science and Technology, Japan Society for the Promotion of Science.
Abstract (Originalsprache)
In the field of data-driven material development, an imbalance in data sets where data points are concentrated in certain regions often causes difficulties in building regression models when machine learning methods are applied. One example of inorganic functional materials facing such difficulties is photocatalysts. Therefore, advanced data-driven approaches are expected to help efficiently develop novel photocatalytic materials even if an imbalance exists in data sets. We propose a two-stage machine learning model aimed at handling imbalanced data sets without data thinning. In this study, we used two types of data sets that exhibit the imbalance: the Materials Project data set (openly shared due to its public domain data) and the in-house metal-sulfide photocatalyst data set (not openly shared due to the confidentiality of experimental data). This two-stage machine learning model consists of the following two parts: the first regression model, which predicts the target quantitatively, and the second classification model, which determines the reliability of the values predicted by the first regression model. We also propose a search scheme for variables related to the experimental conditions based on the proposed two-stage machine learning model. This scheme is designed for photocatalyst exploration, taking experimental conditions into account as the optimal set of variables for these conditions is unknown. The proposed two-stage machine learning model improves the prediction accuracy of the target compared with that of the one-stage model.
Autor:innen und Institutionen
- Division of Materials Science, Nara Institute of Science and Technology, Ikoma-shi, Nara-ken 630-0192, Japan.
- Data Science Center, Nara Institute of Science and Technology, Ikoma-shi, Nara-ken 630-0192, Japan.
- Department of Applied Chemistry, Faculty of Science, Tokyo University of Science, Shinjuku-ku, Tokyo 162-8601, Japan.
- Carbon Value Research Center, Research Institute for Science & Technology, Tokyo University of Science, Noda-shi, Chiba-ken 278-8510, Japan.
- Center for Material Research Platform, Nara Institute of Science and Technology, Ikoma-shi, Nara-ken 630-0192, Japan.
ORCID-Profile: Wataru Takahara, Tomoaki Takayama, Akihiko Kudo, Mikiya Fujii
Förderung
- Ministry of Education, Culture, Sports, Science and Technology
- Japan Society for the Promotion of Science
Zitieren
APA
Takahara, W., Baba, R., Harashima, Y., Takayama, T., Takasuka, S., Yamaguchi, Y., Kudo, A., & Fujii, M. (2025). Data-Driven Approach Considering Imbalance in Data Sets and Experimental Conditions for Exploration of Photocatalysts. ACS Omega, 10(15), 14626-14639. https://doi.org/10.1021/acsomega.4c06997
Vancouver
Takahara W, Baba R, Harashima Y, Takayama T, Takasuka S, Yamaguchi Y, et al. Data-Driven Approach Considering Imbalance in Data Sets and Experimental Conditions for Exploration of Photocatalysts. ACS Omega. 2025;10(15):14626-14639. doi:10.1021/acsomega.4c06997
BibTeX
@article{Takahara2025,
author = {Takahara, Wataru and Baba, Ryuto and Harashima, Yosuke and Takayama, Tomoaki and Takasuka, Shogo and Yamaguchi, Yuichi and Kudo, Akihiko and Fujii, Mikiya},
title = {Data-Driven Approach Considering Imbalance in Data Sets and Experimental Conditions for Exploration of Photocatalysts},
journal = {ACS Omega},
year = {2025},
volume = {10},
number = {15},
pages = {14626--14639},
doi = {10.1021/acsomega.4c06997}
}
EUVIMED