Predicting fertility from seminal traits: Performance of several parametric and non-parametric procedures

Breeding programmes to improve male reproductive performance and efficiency of insemination dose production in paternal lines: feasibility and limitations

“…There are many factors that have an impact on fertility and prolificacy of rabbit does after AI, such as sperm abnormalities and acrosome status (Lavara et al, 2005;Piles et al, 2013a). To our knowledge, only Lavara et al (2012a) have reported h 2 estimates for these traits.…”

“…There is only one research work assessing the predictive ability of male fertility from seminal traits in an independent set of data (i.e., in a data set not used to obtain the predictive function, which allows us to know the ability to predict fertility of future semen samples from a set of explanatory variables measured on them instead of having just an indicator of the goodness of fit of the predictive function to the data used to obtain it; Piles et al, 2013a). In this experiment, AI was performed after a small pre-selection of the ejaculates and 24 h of dose storage at 18°C.…”

“…Promising traits are those that have been reported to have a relevant phenotypic correlation with male reproductive performance: mass motility and number of motile sperm per ejaculate (Bencheikh, 1993;Brun et al, 2002;Theau-Clément et al, 2011); percentage of motile spermatozoa; concentration of spermatozoa in the ejaculate and variables related to it, correlated to the male prolificacy (Brun et al, 2002;TheauClement et al, 2011); pH of the ejaculate (O'Ferral and Meacham, 1968;Vrillon et al, 1979;Bencheikh, 1993;Cofey, 1998, Tusell et al, 2011b; percentage of total motile cells, some sperm movement characteristics measured with CASA systems (linearity index, amplitude of lateral head displacement), percentage of abnormal sperm in the sample (Lavara et al, 2005;Theau-Clément et al, 2011); rate of spermatozoa with presence of cytoplasmatic droplet and rate of reacted spermatozoa during the process of acrosome reaction induction (Piles et al, 2013a).…”

Performance of Classification Techniques on Medical Datasets

“…Predicting ability was measured in terms of rate of erroneous classifications, linear loss (average of the distance between the predicted and the observed classes), the number of predicted classes and the F1 statistic (which allows comparing procedures taking into account that they can predict different number of classes). For all methods, the reduced models showed almost an irrelevant decrease in their predictive abilities compared to the corresponding values obtained with the full models [21].…”

Ordinal classification/regression for analyzing the influence of superstars on spectators in cinema marketing

“…This kind of problem is called ordinal regression or ordinal classification among researchers of the field. There are many problems of this kind recently solved under this paradigm, for instance, for text classification (Baccianella et al, 2013) where documents are classified in an ordinal scale, for predicting fertility rate (Piles et al, 2013) or for analyzing the migrants remitting patterns (Campoy-Muñoz et al, 2014). On the one hand, this problem can be considered as a classification task because the values of the class are in fact labels, although they are commonly represented by numbers.…”