CONICET | Buscador de Institutos y Recursos Humanos

Grouping parent lines into heterotic groups is fundamental in private and public maize breeding programs in order to reduce the number of crosses and field tests necessary to evaluate potential high-yielding hybrids. Molecular markers, such as microsatellites, have been proposed to assign new inbred to heterotic groups previously established with variable results. We conjecture that traditional distance-based methods currently available do not capture the non-linear relation between parental molecular data and progeny performance and that such hindrance can be overcome by supervised learning algorithms. Among them, support vector machines (SVMs) have shown particular high generalization abilities and have become very popular in the last few years. However, they are binary classifiers and a combination scheme is necessary to extend SVMs for problems with more two classes. In this work we explore the performance of the recently introduced class of ECOC-SVM (Error Correcting Output Coding-Support Vector Machine) classifiers, based on recursive error correcting codes of communication theory, in heterotic group assignation. We also report the performance of other ensemble methods reported in the bibliography.