Machine learning-driven microplastics identification using ensemble stacking with Extra Tree meta-models from FTIR data

Khurram Shahzad

doi:10.1016/j.jece.2025.118315

Machine learning-driven microplastics identification using ensemble stacking with Extra Tree meta-models from FTIR data

2025-08-14

0

PRE

AI

DOI：10.1016/j.jece.2025.118315

Original

Original request for help

Save

Abstract

En 中文

• A novel ensemble MLStackXT model is developed to classify microplastics. • Model achieves 95.85% accuracy on Kedzierski, and 95.00% on Jung datasets. • Confusion matrix shows 100% accuracy in 8 of 12 microplastic categories. • It outperforms previous ML and DL models in kappa, F1-score, and accuracy. • The study promotes accurate, transparent plastic pollution monitoring via AI.

Keywords:

microplastics

machine learning

ensemble model

accuracy

plastic pollution monitoring

Machine learning-driven microplastics identification using ensemble stacking with Extra Tree meta-models from FTIR data

Abstract

Journal

Organization