Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Mechanistic Models: Compartment Models in Individual and Population Analysis

Mechanistic Models: Compartment Models in Individual and Population Analysis

Mechanistic models are utilized in individual analysis using single-source data, but imperfections arise due to data collection errors, preventing perfect prediction of observed data. The mathematical equation involves known values (Xi), observed concentrations (Ci), measurement errors (εi), model parameters (ϕj), and the related function (ƒi) for i number of values. Different least-squares metrics quantify differences between predicted and observed values. The ordinary least...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Hereditary spastic paraplegia in three siblings with distinct genetic mutations.

The Journal of international medical research·2026

Same author

Preferential Upregulation of AMOT-p80 Is Associated with YAP-Linked Resistance to 5-Fluorouracil and Oxaliplatin in Colorectal Cancer Cells.

Biomolecules·2026

Same author

Increased mobility, toxicity, and bioaccessibility of arsenate adsorbed onto UV-aged polyvinyl chloride microplastics.

Ecotoxicology and environmental safety·2026

Same author

Functional differentiation of glomalin-related soil protein fractions reveals dual pathways for carbon storage in organic farming systems.

The Science of the total environment·2026

Same author

Deep-eutectic-solvent synthesis of ferrocene-derived Fe-functionalized activated carbon for adsorption- assisted heterogeneous fenton oxidation in complex dye matrices.

Journal of environmental management·2026

Same author

Microbial efficiency enhancement drives carbon sequestration in long-term organic farming systems: linking taxonomic succession to carbon use efficiency.

Frontiers in microbiology·2026

Search research articles

Related Experiment Video

Updated: Sep 15, 2025

Use of Principal Components for Scaling Up Topographic Models to Map Soil Redistribution and Soil Organic Carbon

Use of Principal Components for Scaling Up Topographic Models to Map Soil Redistribution and Soil Organic Carbon

Published on: October 16, 2018

Matrix-specific PFAS source allocation machine learning-based models: Identifying differential indicators in soil and

Jin-Kyung Hong¹, Sungjik Oh², Tae Kwon Lee³

¹Department of Environment and Energy Engineering, Chnonnam National University, Gwangju, 61186, Republic of Korea.

Environmental Research

|July 12, 2025

Summary

New machine learning models accurately identify sources of polyfluoroalkyl substances (PFAS) in soil and water. This matrix-specific approach simplifies PFAS source tracking and reduces analytical costs by identifying key indicator compounds.

Keywords:

Indicator Machine learning PFAS Source allocation Water soil

More Related Videos

Watershed Planning within a Quantitative Scenario Analysis Framework

Watershed Planning within a Quantitative Scenario Analysis Framework

Published on: July 24, 2016

Spatial Multiobjective Optimization of Agricultural Conservation Practices using a SWAT Model and an Evolutionary Algorithm

Spatial Multiobjective Optimization of Agricultural Conservation Practices using a SWAT Model and an Evolutionary Algorithm

Published on: December 9, 2012

Related Experiment Videos

Last Updated: Sep 15, 2025

Use of Principal Components for Scaling Up Topographic Models to Map Soil Redistribution and Soil Organic Carbon

Use of Principal Components for Scaling Up Topographic Models to Map Soil Redistribution and Soil Organic Carbon

Published on: October 16, 2018

Watershed Planning within a Quantitative Scenario Analysis Framework

Watershed Planning within a Quantitative Scenario Analysis Framework

Published on: July 24, 2016

Spatial Multiobjective Optimization of Agricultural Conservation Practices using a SWAT Model and an Evolutionary Algorithm

Spatial Multiobjective Optimization of Agricultural Conservation Practices using a SWAT Model and an Evolutionary Algorithm

Published on: December 9, 2012

Area of Science:

Environmental Chemistry
Environmental Science
Toxicology

Background:

Polyfluoroalkyl substances (PFAS) are persistent environmental contaminants with significant human health risks.
Current PFAS source-allocation methods are costly and error-prone due to extensive data requirements and overlooking matrix-specific behaviors.

Purpose of the Study:

To develop and validate matrix-specific machine-learning classifiers for discriminating PFAS sources (aqueous film-forming foam [AFFF] vs. non-AFFF) in soil and water.
To identify key PFAS indicator compounds for efficient source tracking and reduced analytical burden.

Main Methods:

Compiled a comprehensive dataset of PFAS concentrations from peer-reviewed literature (2012-2024) for soil, water, and AFFF formulations.
Applied fifteen classification algorithms via H2O.AutoML to log-transformed concentrations of 12 legacy PFAS compounds.
Utilized feature-importance analysis and stepwise variable reduction to identify optimal indicator compounds and assess model performance.

Main Results:

Optimal water model (Gradient Boosting Machine) achieved AUC 0.9864; optimal soil model (Distributed Random Forest) achieved AUC 0.9936.
Feature importance identified PFOS, PFHxS, and PFPeS as key water indicators, and PFHxS, PFPeA, and PFOS as key soil indicators.
Source allocation accuracy >0.92 maintained with only 9 PFAS in water and 6 in soil, significantly reducing data needs.

Conclusions:

Matrix-specific machine learning models provide accurate and efficient PFAS source allocation in both soil and water.
Identifying key "sentinel" PFAS indicators substantially reduces analytical requirements without compromising classification performance.
This approach enhances forensic source tracking and informs more efficient environmental remediation strategies.