From rows to yields: how foundation models for tabular data simplify crop yield prediction.
2026Scientific reports
Sabo F, Meroni M, Piles M, Claverie M, Ferreira F +3 more
PubMedDR. FRANK MICHAEL SABO JR., M.D.
LORAIN, OH
Orthopaedic Surgery NPI registered 21+ years 4 publications 2020 – 2026 NPI: 1063417251
AlgorithmsEnvironmental MonitoringDeep LearningClimateCrops, Agricultural
What does FRANK SABO research?
Dr. Sabo studies how advanced computational models, like networks of oscillators and deep learning techniques, can be applied to solve real-world problems. His research involves predicting crop yields for important grains such as barley and wheat, particularly in developing regions where food security is a concern. Additionally, he has created cloud-free satellite imagery that helps classify land use, assisting in understanding environmental changes and land management techniques.
Key findings
- In crop yield forecasting, traditional machine learning methods outperformed deep learning across all tested crops and time periods, mainly due to insufficient data for deep learning.
- The study on oscillators outlines their potential to enhance machine learning applications, thereby improving optimization processes and problem-solving techniques.
- The global map produced from satellite data contains 15 terabytes of information that aids in comprehensive land cover classification, benefiting land management and environmental policy.
Frequently asked questions
- Does Dr. Sabo study crop yields?
- Yes, he focuses on predicting crop yields for grains like barley and durum wheat, especially with limited data.
- What type of technology does Dr. Sabo research?
- He researches advanced computational technologies, including deep learning and networks of oscillators, for agricultural and environmental applications.
- How is Dr. Sabo's work relevant to food security?
- His research helps improve the accuracy of crop yield predictions, which is crucial for addressing food shortages in vulnerable regions.
Publications in plain English
Computing with oscillators from theoretical underpinnings to applications and demonstrators.
2024Npj unconventional computing
Todri-Sanial A, Delacour C, Abernot M, Sabo F
Plain English
This paper studies networks of oscillators, which are systems that can synchronize and work together to solve complex problems. The authors review how these oscillators can be used for various tasks, such as improving machine learning and solving optimization issues. Understanding how oscillator networks work can lead to new technologies and better problem-solving methods.
Who this helps: This benefits researchers and engineers working on advanced computing and artificial intelligence.
Is deeper always better? Evaluating deep learning models for yield forecasting with small data.
2023Environmental monitoring and assessment
Sabo F, Meroni M, Waldner F, Rembold F
Plain English
This study looked at how well advanced deep learning models can predict crop yields in Algeria, focusing on barley, durum wheat, and soft wheat, using limited data from 2002 to 2018. The researchers found that traditional machine learning methods performed better than deep learning for all crops and time periods, mainly because there wasn't enough data for deep learning to work effectively. This matters because understanding crop yield predictions is crucial for food security, especially in vulnerable countries.
Who this helps: This helps farmers and policymakers in food-insecure countries.
A global cloud free pixel- based image composite from Sentinel-2 data.
2020Data in brief
Corbane C, Politis P, Kempeneers P, Simonetti D, Soille P +5 more
Plain English
This research created a detailed, cloud-free map using satellite images from the Sentinel-2 program, covering the world from January 2017 to December 2018. The study produced a huge dataset of 15 terabytes that allows for better land cover classification, making it easier to see and analyze different types of land, like forests or urban areas. This is important because it helps researchers and governments understand how land use is changing over time, which can impact environmental policies and land management.
Who this helps: This helps researchers, environmental scientists, and policymakers.
Frequent Co-Authors
Michele Meroni Felix Rembold María Piles Martin Claverie Fanie Ferreira Elna Van Den Berg Francesco Collivignarelli Aida Todri-Sanial Corentin Delacour Madeleine Abernot
Physician data sourced from the NPPES NPI Registry . Publication data from PubMed . Plain-English summaries generated by AI. Not medical advice.