Application of machine learning and genetic optimization algorithms for modeling and optimizing soybean yield using its component traits
Problem
Predicting total soybean seed yield from yield component traits to support breeding higher-yielding varieties.
Why it matters
Improving soybean yield is presented as the most sustainable way to meet growing global food demand.
What the AI did
MLP, RBF and Random Forest models, individually and via a bagging ensemble, predicted seed yield from five component traits measured on 250 genotypes across four environments.
Results
RBF achieved highest coefficient; exact value truncated in abstract
Impact
Evidence supports ML predicting soybean yield from component traits, but final metrics are cut off.
Record
| Venue | PLoS ONE |
|---|---|
| Published | 2021 |
| Work type | article |
| Authors | Mohsen Yoosefzadeh-Najafabadi, Dan Tulpan, Milad Eskandari |
| Citations | 59 (OpenAlex, at import time) |
| References | 126 |
| Model or method | MLP, RBF, Random Forest, bagged ensemble (E-B) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1371/journal.pone.0250665 |
| Open access | gold |
| Keywords | soybean yield, machine learning, random forest, ensemble, crop breeding |
Importance rating
significant — thousands to millions affected
Abstract frames improving soybean genetic yield potential as the most sustainable way to address growing global food demand and security, tying the problem to global food supply.