Comparing regression, naive Bayes, and random forest methods in the prediction of individual survival to second lactation in Holstein cattle

Abstract

In this study, we compared multiple logistic regression, a linear method, to naive Bayes and random forest, 2 nonlinear machine-learning methods. We used all 3 methods to predict individual survival to second lactation in dairy heifers. The data set used for prediction contained 6,847 heifers born between January 2012 and June 2013, and had known survival outcomes. Each animal had 50 genomic estimated breeding values available at birth and up to 65 phenotypic variables that accumulated over time. Survival was predicted at 5 moments in life: at birth, at 18 mo, at first calving, at 6 wk after first calving, and at 200 d after first calving. The data sets were randomly split into 70% training and 30% testing sets to evaluate model performance for 20-fold validation. The methods were compared for accuracy, sensitivity, specificity, area under the curve (AUC) value, contrasts between groups for the prediction outcomes, and increase in surviving animals in a practical scenario. At birth and 18 mo, all methods had overlapping performance; no method significantly outperformed the other. At first calving, 6 wk after first calving, and 200 d after first calving, random forest and naive Bayes had overlapping performance, and both machine-learning methods outperformed multiple logistic regression. Overall, naive Bayes has the highest average AUC at all decision points up to 200 d after first calving. Random forest had the highest AUC at 200 d after first calving. All methods obtained similar increases in survival in the practical scenario. Despite this, the methods appeared to predict the survival of individual heifers differently. All methods improved over time, but the changes in mean model outcomes for surviving and non-surviving animals differed by method. Furthermore, the correlations of individual predictions between methods ranged from r = 0.417 to r = 0.700; the lowest correlations were at first calving for all methods. In short, all 3 methods were able to predict survival at a population level, because all methods improved survival in a practical scenario. However, depending on the method used, predictions for individual animals were quite different between methods.

Identification of Age-Specific and Common Key Regulatory Mechanisms Governing Eggshell Strength in Chicken Using Random Forests

“…In practice, these SNP effects are unknown and may not even strictly follow a certain distribution [25]. Unlike these traditional statistical models, machine learning methods do not require these prior assumptions about the genetic architecture of traits and have been applied in GWAS in humans [30] as well as in livestock [27,95]. Especially, Romagnoni et al [30] and Huang et al [24] showed that machine learning based algorithms provide promising prediction power to assess genotype-phenotype associations.…”

“…To overcome these limitations of GWAS, application of Bayesian frameworks as well as machine learning algorithms have gained importance in the last decade [21][22][23][24][25]. Their comparative performance has been evaluated for a variety of traits with different genetic architectures (see the reviews [13,26,27]). Nevertheless, multiple studies have revealed that machine learning algorithms surpass currently available well-known GWAS approaches in identifying genes having small effects on the phenotype [28][29][30].…”

Predicting Rice Pest Population Occurrence with Satellite-Derived Crop Phenology, Ground Meteorological Observation, and Machine Learning: A Case Study for the Central Plain of Thailand

“…RF are fast and easy to implement, yields highly accurate predictions and can handle a very large number of input variables without overfitting [49]. This method has been implemented in many different fields in recent years [50] including crop pest and disease prediction. Ayub [51] applied several data mining techniques such as random forest, support vector machine, neural network, k-nearest neighbors, decision tree, and Gaussian naïve bayes to predict grass grub damage and indicated that neural network and random forest performed slightly better than other classifiers.…”

Combining Random Forests and a Signal Detection Method Leads to the Robust Detection of Genotype-Phenotype Associations

“…Multiple studies have confirmed the superiority of machine learning algorithms compared to GWAS approaches by identifying genes having small effects on the phenotype [ 26 , 29 , 30 ]. Machine learning methods do not require prior assumptions about the distribution of the SNP effects, hence can be used for a wide variety of traits in humans [ 31 ], plants [ 28 ] and livestock [ 32 , 33 ]. In particular, Random Forests (RF) models have been praised for their ability to analyze a large number of loci simultaneously and to identify promising associations [ 29 , 30 ].…”