Prediction of Grain Yield in Wheat by CHAID and MARS Algorithms Analyses

Genetic information obtained from ancestral species of wheat and other registered has brought about critical research, especially in breeding, shown great potential for the development advanced breeding techniques. The purpose this study was to determine correlations between some morphological traits various (Triticum spp.) demonstrate application MARS CHAID algorithms wheat-derived data sets. Relationships among several were investigated using a total 26 different genotypes. mining methods compared grain yield prediction cross-validation. In addition, an optimal tree structure with minimum RMSE cross-validated nine terminal nodes. Based on smallest cross-validation, eight-element model found be best prediction. algorithm proved superior accounted 95.7% variation wheats. analyses performed first time research. context, we showed how can help breeders describe complex interaction effects more precisely. With methodology demonstrated study, predict which are beneficial increasing yield. adaption should benefit