1
Return

Cross-domain polymer property prediction by feature engineering using Shapley additive explanations

delete2026-07-07
delete0
PRE
AI
T
Takuya Ehiro *
A
Akiko Nakahashi
DOI：10.1016/j.matlet.2026.141229delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• SHAP-based feature extraction enables cross-domain transfer learning for polymer properties. • Target-domain prediction accuracy improves even with limited training data. • The proposed method outperforms neural network transfer learning on small datasets. • Negative transfer is mitigated by filtering low-importance features.

Journal

Materials Letters cover
Materials Letters
IF:
2.7
Papers:
4.9K
Citations:
5.9W

Organization

No organization information available
Cited Papers

Cited Papers

Citing Papers

Citing Papers