Artificial Neural Network Models as a Decision Support Tool for Selection in Sugarcane: A Case Study Using Seedling Populations

Artificial neural network (ANN) models are mathematical models based on biological neural networks; they are a supervised learning method and use pattern learning from a training dataset that is a subsample of the whole dataset to produce predictions of response variables. We demonstrate the potenti...

Full description

Saved in:
Bibliographic Details
Published in:Crop science 2011, Vol.51 (1), p.21-31
Main Authors: Zhou, Marvellous M, Kimbeng, Collins M, Tew, Thomas L, Gravois, Kenneth A, Pontif, Michael J
Format: Article
Language:English
Subjects:
Agronomy. Soil science and plant productions
artificial selection
Biological and medical sciences
calibration
case studies
Data collection
decision support systems
diameter
Fundamental and applied biological sciences. Psychology
height
Interest rates
mathematical models
Neural networks
plant breeding
Probability
Process controls
Saccharum
seedlings
selection criteria
selection intensity
selection response
Statistical methods
stems
Studies
sugarcane
Variables
yield components
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Artificial neural network (ANN) models are mathematical models based on biological neural networks; they are a supervised learning method and use pattern learning from a training dataset that is a subsample of the whole dataset to produce predictions of response variables. We demonstrate the potential of an ANN model as a tool for selection in sugarcane. Cane yield components, namely stalk number, stalk height, and stalk diameter, were measured on individual seedlings and used as predictor variables to produce a selection decision (reject or select a seedling) based on an ANN model. Compared with the currently used visual method of selection, the difference in cane yield between the mean of the selected and rejected seedlings was greater for seedlings selected by the ANN model. The difference increased when similar selection intensity was applied in both selection methods. The ANN model selected fewer seedlings with cane yield lower than the population mean and rejected fewer seedlings with higher cane yield compared with the visual method. The ANN model compels the breeder to consider all traits simultaneously when deciding whether to select or reject a clone, which is likely to be more efficient than judging the merit by considering each trait independently. The ANN model can be a valuable tool to determine selection rates to be applied in selecting sugarcane families during seedling selection.
ISSN:0011-183X
1435-0653
DOI:10.2135/cropsci2010.02.0057