The Use of Selected Machine Learning Methods in Dairy Cattle Farming: A Review

Area of Science:

• Agricultural Science
• Computer Science
• Data Science

Background:

• Dairy cattle farming increasingly utilizes advanced computational methods.
• Machine learning (ML) offers powerful tools for optimizing dairy operations.
• Recent advancements necessitate a review of ML applications in this sector.

Purpose of the Study:

• To review selected machine learning algorithms used in dairy cattle farming between 2020 and 2024.
• To describe the application of various ML methods in dairy cattle breeding and husbandry.
• To outline model construction, implementation, and performance evaluation for ML models.

Main Methods:

• Review of machine learning algorithms including regression, trees, random forest, AdaBoost, SVM, k-NN, Naive Bayes, MARS, ANNs (DNNs, CNNs), GMMs, and cluster analysis.
• Description of model construction, implementation stages, and performance indicators.
• Analysis of time trends in ML method popularity within dairy cattle farming.

Main Results:

• Selected ML algorithms like random forest, ANNs, and SVMs show significant application in dairy cattle farming.
• Examples provided cover diverse areas such as breeding, health, and production management.
• Performance indicators for regression and classification models are detailed, aiding in method selection.

Conclusions:

• Machine learning is a rapidly evolving and increasingly vital tool in modern dairy cattle farming.
• Understanding different ML algorithms and their applications is crucial for effective farm management.
• The trend indicates continued growth and integration of ML in optimizing dairy production and animal welfare.