Application of Machine Learning Algorithms to Describe the Characteristics of Dairy Sheep Lactation Curves

SIMPLE SUMMARY: An accurate estimation of the characteristics lactation curves is required to optimize sheep milk production. The adjustment of the lactation curve is traditionally been performed using mathematical models through linear and non-linear regression. However, these analytical tools have several limitations, mainly related to the non-linear pattern of the lactation curve. Machine learning algorithms have been used successfully to model and predict complex biological processes. In the current study, we evaluated the ability of seven machine learning algorithms, including linear and non-linear regression, to estimate total milk yield, peak yield, and time to peak yield of dairy sheep lactations. In addition, the estimates provided by machine learning algorithms were compared with the Wood model and the observed values. All algorithms tested showed good estimates, with the SMOreg algorithm showing the best performance. Furthermore, our results indicated that adequate estimates can be obtained with only five milk records. Therefore, machine learning algorithms are an option to correctly predict the characteristics of the lactation curve of dairy sheep, optimizing the use of available data. ABSTRACT: In recent years, machine learning (ML) algorithms have emerged as powerful tools for predicting and modeling complex data. Therefore, the aim of this study was to evaluate the prediction ability of different ML algorithms and a traditional empirical model to estimate the parameters of lactation curves. A total of 1186 monthly records from 156 sheep lactations were used. The model development process involved training and testing models using ML algorithms. In addition to these algorithms, lactation curves were also fitted using the Wood model. The goodness of fit was assessed using correlation coefficient (r), mean absolute error (MAE), root mean square error (RMSE), relative absolute error (RAE), and relative root mean square error (RRSE). SMOreg was the algorithm with the best estimates of the characteristics of the sheep lactation curve, with higher values of r compared to the Wood model (0.96 vs. 0.68) for the total milk yield. The results of the current study showed that ML algorithms are able to adequately predict the characteristics of the lactation curve, using a relatively small number of input data. Some ML algorithms provide an interpretable architecture, which is useful for decision-making at the farm level to maximize the use of available information.