Real-time milk analysis integrated with stacking ensemble learning as a tool for the daily prediction of cheese-making traits in Holstein cattle

Abstract

Cheese-making traits in dairy cattle are important to the dairy industry but are difficult to measure at the individual level because there are limitations on collecting phenotypic information. Mid-infrared spectroscopy has its advantages, but it can only be used during monthly milk recordings. Recently, in-line devices for real-time analysis of milk quality have been developed. The AfiLab recording system (Afimilk) offers significant benefits as phenotypes can be collected from each cow at each milking session. The objective of this study was to assess the potential of integrating AfiLab real-time milk analyzer measures with the stacking ensemble learning technique using heterogeneous base learners for the in-line daily monitoring of cheese-making traits in Holstein cattle with a view to developing a precision livestock farming system for monitoring the technological quality of milk. Data and samples for wet-laboratory analyses were collected from 499 Holstein cows belonging to 2 farms where the AfiLab system was installed. The traits of concern were 9 milk coagulation traits [3 milk coagulation properties (MCP), and 6 curd firming traits (CF t )], and 7 cheese-making traits [3 cheese yield (CY) traits, and 4 milk nutrient recovery in the curd (REC) traits]. The near-infrared AfiLab spectral data and on-farm information (days in milk and parity) were used to assess the predictive ability of different statistical methods [elastic net (EN), gradient boosting machine (GBM), extreme gradient boosting (XG-Boost), and artificial neural network (ANN)] across different cross-validation scenarios. These statistical methods were considered the base learners, which were then combined in a stacking ensemble learning. Results indicate that including information on the cows (days in milk and parity) in the AfiLab infrared prediction increased its accuracy by 10.3% for traditional MCP, 13.8% for curd firming, 9.8% for CY, and 11.2% for REC traits compared with those obtained from nearinfrared AfiLab alone. The statistical approaches exhibited high prediction accuracies (R 2 ) averaged across the cross-validation scenarios for traditional MCP (0.58 for ANN, 0.55 for EN and GBM, 0.52 for XGBoost, and 0.62 for stacking ensemble), CF t (0.55 for ANN, 0.54 for EN and GBM, 0.53 for XGBoost, and 0.61 for stacking ensemble), and similar R 2 averages for CY and REC (0.55 for ANN, 0.54 for EN and GBM, 0.53 for XGBoost, and 0.61 for stacking ensemble). The ANN approach was more accurate than the other base learners (EN, GBM, and XGBoost) and improved accuracy across cross-validation scenarios on average by 7% for traditional MCP, 5% for CF t , 8% for CY, and 7% for REC. The stacking ensemble method improved prediction accuracy by 3% to 31% for traditional MCP, 2% to 26% for CF t , 1% to 38% for CY traits, and 2% to 27% for REC traits compared with the base learners. The prediction accuracies of the different approaches evaluated tended to decrease from the 10-fold crossvalidation to the independent validation scenario, although there was...

In-line near-infrared analysis of milk coupled with machine learning methods for the daily prediction of blood metabolic profile in dairy cattle

“…feed, management). Indeed, the inclusion of DIM and parity effects has been reported to enhance the robustness of prediction equations 22 , 53 . Here, to examine the potential of NIR analysis, we built prediction equations on the basis of a single breed and a single herd, and included in the models the main sources of variation in lactating cows (i.e.…”

“…Near-infrared (350 to 2500 nm; NIR) spectra, on the other hand, appear to be just as reliable as FTIR spectra for analyzing raw milk composition and cheese-making traits 20 – 22 . The visible to low-NIR range (350 to 1000 nm) seems to be suitable for automated in-line analysis of milk, as the optical sensors are less expensive 23 .…”

Hyperspectral technique combined with stacking and blending ensemble learning method for detection of cadmium content in oilseed rape leaves

“…proved that ensemble learning model is superior to the basic model in predicting the yield of individual apple trees using an unmanned aerial vehicle. Mota et al 33 . proved that the stacking ensemble learning model has a stronger prediction ability for cheesemaking traits.…”

“…Chen et al 32 proved that ensemble learning model is superior to the basic model in predicting the yield of individual apple trees using an unmanned aerial vehicle. Mota et al 33 proved that the stacking ensemble learning model has a stronger prediction ability for cheesemaking traits. It is also proved that the prediction accuracy of stacking models was slightly higher than that of blending models in this paper.…”

Comparison of machine learning models for bluetongue risk prediction: a seroprevalence study on small ruminants

“…Gradient boosting machine and ANN outperformed RF, stacking clusters, and other regression models among the techniques used. In a study conducted by Mota et al [ 15 ] to predict cheese-making traits in Holstein cattle, they compared the prediction of ANN, elastic net, gradient boosting machine, and extreme gradient boosting, and ANN achieved the highest predictions.…”