Use of multivariate adaptive regression splines for prediction of body weight from body measurements in Marecha (Camelus dromedaries) camels in Pakistan

Using Multivariate Adaptive Regression Splines to Estimate the Body Weight of Savanna Goats

“…The predictive performance of the MARS data mining algorithm of the current study showed that root mean square error, relative root mean square error, standard deviation ratio, coefficient of variation, performance index, mean error, relative approximation error, mean absolute percentage error, mean absolute deviation and alkaike’s information criterion were lower in the training dataset than test, whereas Pearson’s correlation coefficient, coefficient of determination and adjusted coefficient of determination were higher in the training dataset than the test. The Pearson’s correlation coefficient and coefficient of determination for this study were higher than the study by Tirink et al [ 32 ], which highlighted that the model with 80:30 proportions was the best model in the Marercha camel of Pakistan with the MARS model that resulted in eight basic functions. Faraz et al [ 33 ] reported a lower goodness of fit for the prediction of live body weight in Thalli sheep, and Tyasi et al [ 34 ] reported a lower goodness of fit for the prediction of body weight of the Hy-Line Silver Brown commercial layer chicken breed.…”

Use of Multivariate Adaptive Regression Splines Algorithm to Predict Body Weight from Body Measurements of Anatolian buffaloes in Türkiye

“…Faraz et al [ 37 ] compared CART and MARS algorithms to predict live body weight based on body measurements in Thalli sheep; they mentioned that the MARS algorithm was superior to CART according to the comparison criteria used. Tırınk et al [ 10 ] used the MARS algorithm to predict body weight from body measurements in Marecha (Camelus dromedaries) camels. They mentioned that the best MARS model for BW prediction was obtained using sex and shoulder height as independent variables for an 80:20 training and test set proportion.…”

“…Live weight estimations based on body measurements in buffaloes can reveal growth and development characteristics as well as production performance and genetic characteristics of buffaloes. In recent years, several body measurements taken during early growth periods have been used as early selection criteria to improve the relative proportion of superior buffalo offspring with good body weight in future populations [ 10 ]. It has also been reported that these measurements are practically helpful for buffalo breeders willing to estimate body weight, which is essential for herd management.…”

“…In rural conditions where there is no scale, buffalo body measurements and body weight estimation contribute to the determination of the optimum feed amount per buffalo in the herd, the marketing price, the dose of medicinal drug to be used in the treatment of diseases and the optimum slaughter time [ 10 ]. In this context, estimation based on body measurements, which is used as one of the positive effects on live weight, is accepted as an indirect selection criterion in buffalo breeding.…”

“…Many studies have been published on estimating body weight from body measurements in different animal species, such as sheep, cattle, rabbits, dogs and camels [ 10 , 11 , 12 , 13 , 14 ]. However, there is no such study for the Anatolian buffalo breed.…”