首页|Investigators at Iowa State University Report Findings in Machine Learning (Leve raging Soil Mapping and Machine Learning To Improve Spatial Adjustments In Plant Breeding Trials)
Investigators at Iowa State University Report Findings in Machine Learning (Leve raging Soil Mapping and Machine Learning To Improve Spatial Adjustments In Plant Breeding Trials)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Machine Learning is th e subject of a report. According to news reporting originating from Ames, Iowa, by NewsRx correspondents, research stated, "Spatial adjustments are used to impr ove the estimate of plot seed yield across crops and geographies. Moving means ( MM) and P-Spline are examples of spatial adjustment methods used in plant breedi ng trials to deal with field heterogeneity." Financial support for this research came from North Central Soybean Research Pro gram. Our news editors obtained a quote from the research from Iowa State University, "Within the trial, spatial variability primarily comes from soil feature gradien ts, such as nutrients, but a study of the importance of various soil factors inc luding nutrients is lacking. We analyzed plant breeding progeny row (PR) and pre liminary yield trial (PYT) data of a public soybean breeding program across 3 ye ars consisting of 43,545 plots. We compared several spatial adjustment methods: unadjusted (as a control), MM adjustment, P-spline adjustment, and a machine lea rning-based method called XGBoost. XGBoost modeled soil features at: (a) the loc al field scale for each generation and per year, and (b) all inclusive field sca le spanning all generations and years. We report the usefulness of spatial adjus tments at both PR and PYT stages of field testing and additionally provide ways to utilize interpretability insights of soil features in spatial adjustments. Ou r work shows that using soil features for spatial adjustments increased the rela tive efficiency by 81%, reduced the similarity of selection by 30%,and reduced the Moran's I from 0.13 to 0.01 on average across all experiments. These results empower breeders to further refine selection criteria to make mor e accurate selections and select for macro- and micro-nutrients stress tolerance . Spatial adjustments utilizing soil maps perform better than traditional method s for spatial adjustments of trials. Soil-based spatial adjustments can be used to better understand the spatial variability in breeding trials. Site-specific m achine learning models for spatial adjustments perform better than large general ized models. Plant breeding trials are a key component of crop improvement for y ield, quality, and stress resistance. Breeding trials typically are grown on sma ll plots of land and are highly affected by the area in the field where they are planted due to field trends. We investigated if using the soil features in a fi eld could explain some of the variability in the early stages of a breeding prog ram and used machine learning techniques to estimate the soil effects on observe d yields. We found that by using the soil features for spatial adjustments, we c ould increase the accuracy of selections and improve the outcomes of decisions m ade by a breeder."
AmesIowaUnited StatesNorth and Cen tral AmericaCyborgsEmerging TechnologiesMachine LearningIowa State Unive rsity
NETL
NSTL
