This study developed predictive models for deoxynivalenol (DON) contamination in spring oats in Sweden using machine learning. Three models were created for farmers, crop collectors, and food safety inspectors. Data included weather, agronomical, site-specific factors, and previous year's DON levels. Random Forest models achieved >0.72 accuracy, with June predictions showing promise. Rainfall, humidity, wind speed, crop variety, and elevation were key predictive features.
Publisher
npj Science of Food
Published On
Oct 04, 2024
Authors
X. Wang, T. Borjesson, J. Wetterlind, H. J. van der Fels-Klerx
Tags
deoxynivalenol
DON contamination
spring oats
machine learning
predictive models
Sweden
agronomical factors
Related Publications
Explore these studies to deepen your understanding of the subject.
