Bayesian optimisation for interval selection in PLS models

doi:10.1016/j.chemolab.2025.105541

Bayesian optimisation for interval selection in PLS models

2025-10-10

0

PRE

AI

DOI：10.1016/j.chemolab.2025.105541

Original

Original request for help

Save

Abstract

En 中文

• A Bayesian optimisation framework is introduced for adaptive interval selection in PLS. • The method replaces exhaustive grid search with an efficient, data-driven approach. • Simultaneously selects single or multiple intervals to improve model performance. • Achieves competitive or superior performance compared to full-spectrum and stepwise iPLS models on real-world NIR datasets. • A Monte Carlo study confirms the algorithm’s robustness and efficient convergence.

Bayesian optimisation for interval selection in PLS models

Abstract

Journal

Organization