“…A number of studies based on traditional risk factors identification approaches (logistic regression mainly) have been performed on BVDV [4-8], and the knowledge about major risk factors are related to the following: biosecurity [6], reproduction management [2,6,9,10], herd size [5,8], animal introduction [2,4,5,11], direct contact with other animals (from the same species or not) [4,11-13], communal grazing [4,5], age of animals [5,14], artificial insemination (AI) [15], and natural mating [13]. Nonetheless, usual epidemiologic analytic frameworks like logistic regression are often limited for the analysis of high-dimensional, imbalanced and nonlinear data, and may be poorly adapted to epidemiological datasets with a large number of predictor variables (parameters) in relation to the number of observations given the high susceptibility to overfitting [16,17].…”