Sensor-based cattle monitoring is a growing field within precision livestock farming, where machine learning algorithms are used to detect health issues, monitor behavior, and improve productivity. However, the diversity of evaluation strategies, performance metrics, and algorithmic choices in the current literature complicates comparison and selection of machine learning models. In this study, we conduct a systematic review and quantitative analysis of machine learning approaches applied to sensor-driven cattle monitoring tasks. We evaluate and compare algorithms using a broad spectrum of performance indicators, including accuracy, F1-score, precision, and recall, alongside their variability across studies. Through detailed heatmap visualizations and statistical summaries (mean, median, standard deviation, coefficient of variation), we identify the most consistently high-performing models and the factors contributing to variability. Results show that Convolutional Neural Networks and Extreme Gradient Boosting achieve the highest F1-scores and most balanced precision-recall performance, especially in imbalanced health detection tasks, while tree ensemble models such as Random Forest and Gradient Boosting demonstrate superior accuracy and consistency for behavior classification on balanced datasets. Our findings emphasize the importance of multi-metric evaluation frameworks and context-aware algorithm selection in the development of scalable and reliable livestock monitoring systems.

Managing reproductive activity in dairy farms is an important factor in increasing cow productivity and making the farm more economically efficient. The research was carried out in the 5 most important Holstein dairy farms in Romania, during the period 2022-2024. The data obtained were processed and interpreted statistically, compared with the results obtained for these parameters by other researchers. Following the research, average values of milk quantity were recorded ranging from 9749.38 ± 20.88 kg of milk in farm D in 2022 and 15514.01 ± 29.31 kg of milk in farm A in 2024. The age of first calving decreased from one year to the latest in each farm, with the averages falling in the last year between 22.53 months and 24.61 months, and the calving interval decreased with an average value on all farms in 2024 of 401 days. The average duration of exploitation of dairy cows is 4 lactations, and based on the calculation of the efficiency index of reproductive activity, the 5 farms are efficient in terms of using cows for reproduction.

In the context of modern ruminant nutrition, increasing attention is being directed toward the valorization of agro-industrial by-products as alternative feed ingredients that enhance nutrient utilization efficiency while supporting the sustainability of animal production systems. Milk thistle (Silybum marianum) oilseed cake, a by-product of oil extraction, has emerged as a resource of growing interest due to its favorable nutritional profile and the presence of bioactive compounds with functional properties. This review critically analyzes recent scientific literature addressing the use of milk thistle oilseed cake in ruminant nutrition, highlighting its potential practical relevance as a functional feed ingredient. The available evidence suggests that milk thistle oilseed cake may support inclusion in ruminant diets at moderate levels; however, controlled in vivo studies remain limited, and several proposed mechanisms are inferred from studies on structurally analogous polyphenol-rich by-products rather than from milk thistle cake itself. Further research is needed before precise inclusion recommendations can be established. Special attention is given to the bioactive fraction dominated by the silymarin complex, which may interact with rumen digestive and fermentative processes, influencing nutrient utilization efficiency and oxidative stability. Overall, the findings suggest that milk thistle oilseed cake represents a promising feed resource that aligns with sustainable and efficiency-oriented feeding strategies in modern ruminant production systems.

Accurate estimation of genetic parameters is a prerequisite for efficient genetic improvement in dairy cattle. In the Romanian Black and White (BNR) breed, the genetic covariance structure among milk production traits has not been simultaneously estimated under a Bayesian multi-trait framework. A Bayesian multi-trait animal model implemented via Markov chain Monte Carlo (MCMC) methods using MCMCglmm was applied to 8207 standard 305-day lactation records from 16,658 animals with pedigree information. Three traits were analyzed simultaneously: milk yield, fat content, and protein content. Heritabilities, genetic, phenotypic, and environmental correlations, and mean accuracy of selection were estimated, with uncertainty quantified through 95% highest posterior density (HPD) intervals. Heritability estimates were h² = 0.249 (95% HPD: 0.172–0.332) for milk yield, h² = 0.504 (95% HPD: 0.380–0.607) for fat content, and h² = 0.383 (95% HPD: 0.284–0.491) for protein content, with compositional traits showing stronger genetic determination than milk yield. Milk yield was negatively correlated with both fat content (r_g= −0.496) and protein content (r_g = −0.574), while fat and protein contents were positively correlated (r_g = 0.643). The results provide a quantitative basis for the BNR national breeding program and confirm that the negative genetic correlations between milk yield and compositional traits should be considered in any selection index targeting simultaneous improvement in production volume and milk quality.