A Comparative Study for Egg Production in Layers by Decision Tree Analysis
A Comparative Study for Egg Production in Layers by Decision Tree Analysis
Yakut Gevrekçi* and Çiğdem Takma
ABSTRACT
Decision tree analysis is one of the data mining approaches which generate model asses the response variable based on the predictor variables. The present study deals with evaluating relationship between egg production (EP) and some related traits by decision tree analysis with Classification and Regression Tree (CART) and Chi-squared Automatic Interaction Detection (CHAID) algorithms. A total of 1912 EP records of layers which hatched at four different periods were collected from a commercial flock in west of Turkey. The analysis performed with cathegorical (cage and line) and continuous (hatch, age at sexual maturity and body weight of sexual maturity) predictors and a continuous response variable (EP). The Spearman’s rho correlation coefficients among all predictor variables were significantly changed from 0.88 to -0.65. Age of sexual maturity was found the most predictor variable that affecting EP and it was followed by hatching period. Lower values for goodness of fit criteria were obtained by CHAID algorithm, indicating that this algorithm fits better. Regression tree analysis displayed visual information about the relations among predictor variables. As a result, decision tree analysis is necessary to provide preliminary information in breeding programs to be performed.
To share on other social networks, click on any share button. What are these?
April 2018
Vol. 50, Iss. 2, Pages 401-797
