論文 Bayesian inference for peak feature extraction and prediction of material property in X-ray diffraction data

Ryo Murakami SAMURAI ORCID (National Institute for Materials Science) ; Taisuke T. Sasaki SAMURAI ORCID (National Institute for Materials Science) ; Hideki Yoshikawa SAMURAI ORCID (National Institute for Materials Science) ; Yoshitaka Matsushita SAMURAI ORCID (National Institute for Materials Science) ; Keitaro Sodeyama SAMURAI ORCID (National Institute for Materials Science) ; Tadakatsu Ohkubo SAMURAI ORCID (National Institute for Materials Science) ; Hiroshi Shinotsuka SAMURAI ORCID (National Institute for Materials Science) ; Kenji Nagata SAMURAI ORCID (National Institute for Materials Science)

Bayesian inference for peak feature extraction and prediction of material property in X-ray diffraction data.pdf
Download file (7.31MB)
Download as zip (6.82MB)
コレクション

引用
Ryo Murakami, Taisuke T. Sasaki, Hideki Yoshikawa, Yoshitaka Matsushita, Keitaro Sodeyama, Tadakatsu Ohkubo, Hiroshi Shinotsuka, Kenji Nagata. Bayesian inference for peak feature extraction and prediction of material property in X-ray diffraction data. Science and Technology of Advanced Materials: Methods. 2024, 4 (1), 2384352. https://doi.org/10.1080/27660400.2024.2384352

説明:

(abstract)

To advance the development of materials through data-driven scientific methods, appropriate methods for building machine learning (ML)-ready feature tables from measured and computed data must be established. In materials development, X-ray diffraction (XRD) is an effective technique for analysing crystal structures and other microstructural features that have information that can explain material properties. Therefore, the fully automated extraction of peak features from XRD data without the bias of an analyst is a significant challenge. This study aimed to establish an efficient and robust approach for constructing peak feature tables that follow ML standards (ML-ready) from XRD data. We challenge peak feature extraction in the situation where only the peak function profile is known a priori, without knowledge of the measurement material or crystal structure factor. We utilized Bayesian estimation to extract peak features from XRD data and subsequently performed Bayesian regression analysis with feature selection to predict the material property. The proposed method focused only on the tops of peaks within localized regions of interest (ROIs) and extracted peak features quickly and accurately. This process facilitated the rapid extracting of major peak features from the XRD data and the construction of an ML-ready feature table. We then applied Bayesian linear regression to the maximum energy product (𝐵𝐻)𝑚𝑎𝑥, using the extracted peak features as the explanatory variable. The outcomes yielded reasonable and robust regression results. Thus, the findings of this study indicated that 004 peak height and area were important features for predicting (𝐵𝐻)𝑚𝑎𝑥.

権利情報:

キーワード: Materials informatics, Spectral decomposition, Bayesian estimation, Feature selection, AI-ready

刊行年月日: 2024-12-31

出版者: Informa UK Limited

掲載誌:

  • Science and Technology of Advanced Materials: Methods (ISSN: 27660400) vol. 4 issue. 1 2384352

研究助成金:

  • MEXT Program: Data Creation and Utilization-Type Material Research and Development Project JPMXP1122715503

原稿種別: 出版者版 (Version of record)

MDR DOI:

公開URL: https://doi.org/10.1080/27660400.2024.2384352

関連資料:

その他の識別子:

連絡先:

更新時刻: 2025-11-10 16:30:53 +0900

MDRでの公開時刻: 2025-11-10 16:25:11 +0900

ファイル名 サイズ
ファイル名 Bayesian inference for peak feature extraction and prediction of material property in X-ray diffraction data.pdf (サムネイル)
application/pdf 		サイズ 7.31MB 詳細