Comparison of artificial neural network accuracy and linear regression model in estimating the weight of dromedary camel using body measurements

In the management of camel breeding, weightlifting plays a decisive role in grouping livestock, regulating nutritional needs and also annual evaluation of animals. Due to the many difficulties and risks, camel owners usually prefer alternative methods such as using apparent estimates or weight -meter to estimate the weight of camels. Since the accuracy of mathematical models in estimating the weight of camels is not equal, so the present research was conducted to compare the accuracy of estimating artificial neural network and multiple linear regression model in estimating the weight of dromedary camels from their body dimensions. For this purpose, 26 camels with 203 records were used from a private farm for one year. Weighing and determining the body dimensions of camels (body length, shoulder height, back height, hump height to the ground, chest and abdomen girth) were measured monthly. To estimate the weight of camels from their body dimensions, the data were analyzed using multiple linear regression model and artificial neural network. The weight of camels on their body dimensions was estimated with accuracy of 0.94 and 0.99, respectively, using multivariate linear regression model and artificial neural network model. The weight of camels on their body dimensions was estimated with accuracy of 0.94 and 0.99, respectively, using multivariate linear regression model and artificial neural network model. The results of this research showed that the artificial neural network has the proper ability to estimate the weight of camels based on their body dimensions and can replace conventional regression methods.