Weilin Yuan
;
Yusuke Hibi
(National Institute for Materials Science
)
;
Ryo Tamura
(National Institute for Materials Science
)
;
Masato Sumita
;
Yasuyuki Nakamura
(National Institute for Materials Science
)
;
Masanobu Naito
(National Institute for Materials Science
)
;
Koji Tsuda
(National Institute for Materials Science
)
Description:
(abstract)The efficient treatment of polymer waste is a major challenge to marine sustainability. It is useful to reveal the factors that dominate the degradability of polymer materials for developing new polymer materials in the future. In this study, we have developed a platform for evaluating the degradability of polymers based on machine learning techniques. However, the small number of available datasets on degradability and the diversity of experimental means and conditions hinder large-scale analysis. To avoid this difficulty, we have introduced RankSVM, which can learn the preference of the degradability of polymers. We have made a ranking model to evaluate the degradability of polymers, integrating three datasets on the degradability of polymers that are measured by different means and conditions. The analysis of this ranking model using a decision tree has revealed the factors that dominate the degradability of polymers.
Rights:
Keyword: polymer degradation, rank-based machine learning, PoLyInfo
Date published: 2023-09-25
Publisher: Elsevier BV
Journal:
Funding:
Manuscript type: Publisher's version (Version of record)
MDR DOI:
First published URL: https://doi.org/10.1016/j.patter.2023.100846
Related item:
Other identifier(s):
Contact agent:
Updated at: 2024-01-05 22:12:08 +0900
Published on MDR: 2023-11-10 13:30:10 +0900
Filename | Size | |||
---|---|---|---|---|
Filename |
1-s2.0-S2666389923002258-main.pdf
(Thumbnail)
application/pdf |
Size | 3.21 MB | Detail |