publications

2025

1. Under Review

   Strategic Nitrogen Recommendations for Potatoes: Soil- and Climate-Specific Insights from the Model-Based Identifying NEMO Framework

   Moreteza Mesbah, Kristen Murchison, Mariaelisa Polsinelli, and 2 more authors

   Agricultural Science & Technology, 2025

   Bib

   @article{mesbahStrategicNitrogenRecommendations2025,
     title = {Strategic Nitrogen Recommendations for Potatoes: Soil- and Climate-Specific Insights from the Model-Based Identifying NEMO Framework},
     author = {Mesbah, Moreteza and Murchison, Kristen and Polsinelli, Mariaelisa and Jégo, Guillaume and Morissette, René},
     year = {2025},
     journal = {Agricultural Science & Technology},
     publisher = {American Chemical Society},
     selected = true
   }

2024

1. Proceedings

   

   Imbalanced Datasets and Crop Yield Prediction: Application of Preprocessing Techniques for Regression Tasks in Agriculture

   Mariaelisa Polsinelli, Morteza Mesbah, Zhiming Qi, and 1 more author

   In 2024 ASABE Annual International Meeting, 2024

   Abs DOI Bib

   Machine learning (ML) is increasingly used in the agricultural sector to predict crop yields. The ability to produce short-term seasonal forecasting or future projections is highly important for creating climate change adaptation strategies. The escalation of extreme weather events such as drought due to climate change is a significant threat to crop yields. While ML algorithms like Random Forest (RF) have shown great performance on balanced datasets, the increase in weather extremes presents a challenge as historical data may lack instances of low yield years, leading to unbalanced datasets biased towards ‘normal’ or ‘well performing’ years. Three preprocessing techniques; random undersampling (US), random oversampling (OS), and the Synthetic Oversampling Technique for Regression (SMOTER) were applied to the training datasets of nine industrial farm field-years growing Russet Burbank using RF. Assessing the nine field-years overall, US improved the relative root mean square error (RRMSE) from the baseline the most reducing it from 16.7[PM(1.1]% to 14.1%. Individually, US was the winning technique for five of the nine fields, likely due to a reduction in overfitting. SMOTER provided the largest improvement in performance for Field 3b (drought year). The baseline RRMSE for this field was 36.8% (poor). The application of SMOTER to synthetically increase low yield data points and high GDD low precipitation values in the training dataset reduced the RRMSE to a satisfactory 26.3%. The results indicate that SMOTER, potentially in combination with other performance techniques, can be a viable option for improving the yield prediction of drought and extreme weather years.

   @inproceedings{polsinelliImbalancedDatasetsCrop2024,
     title = {Imbalanced Datasets and Crop Yield Prediction: Application of Preprocessing Techniques for Regression Tasks in Agriculture},
     booktitle = {2024 ASABE Annual International Meeting},
     author = {Polsinelli, Mariaelisa and Mesbah, Morteza and Qi, Zhiming and Ramsay, Matt},
     year = {2024},
     pages = {1},
     publisher = {American Society of Agricultural and Biological Engineers},
     doi = {10.13031/aim.202400813},
     url = {https://elibrary.asabe.org/abstract.asp?aid=54797},
   }