EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Role of High Fidelity Vs. Low Fidelity Experimental Data in Machine Learning Model Performance for Predicting Polymer Solubility

Mona Amrihesari iD, Manali Banerjee, Raul Olmedo, Blair Brettmann iD

Macromolecular Rapid Communications · 2026

DOI: 10.1002/marc.202500454 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Macromolecular Rapid Communications, 2026. Verfasst von Mona Amrihesari und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, Georgia, USA. Hauptthemen laut MEDLINE: Polymers, Machine Learning. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Office of Naval Research.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Reliable classification of polymer–solvent compatibility is essential for solution formulation and materials discovery. Applying machine learning (ML) and artificial intelligence to this task is of growing interest in polymer science, but the effectiveness of such models depends on the quality/nature of the training data. This study evaluates how experimental data fidelity, as set by the experimental method, influences ML model performance by comparing classifiers trained on two experimental datasets: one generated from turbidity‐based measurements using a Crystal16 parallel crystallizer as a high‐fidelity source and another derived from visual solubility inspection as a low‐fidelity dataset. Both datasets were encoded using one‐hot encoding for polymers and Morgan fingerprints for solvents and modeled using XGBoost classifiers to predict solubility labels as soluble, insoluble, and partially soluble. Confusion matrices showed that models trained on high‐fidelity data better captured partially soluble behavior and more clearly distinguished between classes, highlighting the advantage of quantitative measurements over subjective classification. We also found that adding temperature as a feature improved prediction accuracy for the low‐fidelity dataset—a key consideration for literature‐derived data, which often lacks this information. These findings underscore the importance of experimental rigor and completeness when developing generalizable ML‐based tools for polymer solubility prediction.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PolymersMachine LearningSolubilityPredictive Learning ModelsBoosting Machine Learning AlgorithmsSolventsPrediction AlgorithmsClassification AlgorithmsTemperature

Stichwörter der Autor:innen

datasetmachine learningpredictionsolubility

Substanzen

PolymersSolvents

Autor:innen und Institutionen

  • School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, Georgia, USA.
  • School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia, USA.

ORCID-Profile: Mona Amrihesari, Blair Brettmann

Förderung

  • Office of Naval Research

Zitieren

APA

Amrihesari, M., Banerjee, M., Olmedo, R., & Brettmann, B. (2026). Role of High Fidelity Vs. Low Fidelity Experimental Data in Machine Learning Model Performance for Predicting Polymer Solubility. Macromolecular Rapid Communications. https://doi.org/10.1002/marc.202500454

Vancouver

Amrihesari M, Banerjee M, Olmedo R, Brettmann B. Role of High Fidelity Vs. Low Fidelity Experimental Data in Machine Learning Model Performance for Predicting Polymer Solubility. Macromolecular Rapid Communications. 2026. doi:10.1002/marc.202500454

BibTeX
@article{Amrihesari2026,
  author = {Amrihesari, Mona and Banerjee, Manali and Olmedo, Raul and Brettmann, Blair},
  title = {Role of High Fidelity Vs. Low Fidelity Experimental Data in Machine Learning Model Performance for Predicting Polymer Solubility},
  journal = {Macromolecular Rapid Communications},
  year = {2026},
  doi = {10.1002/marc.202500454}
}
Kostenlos starten →