Table 4.

Application of different ML models through different data and species.

Data type	ML model	Modeling output	Species
			Ruminants	Pigs	Poultry
Production records, feed intake, phenotypic traits.	regularizing models	Production and genomic performance prediction	(44–48, 81, 82)	(49–55)	(56, 83–87)
Sensor data, farm records, health, and management data	Tree-based models	Condition estimation and behavior classification, risk monitoring, and assessment	(88–91)	(92–97)	(98–102)
Accelerometer data, sample spectra, and imaging features.	Support Vector Machines (SVM)	Animal status, carcass, and body condition scoring	(103–108)	(49, 53, 55, 109–112)	(100, 113–116)
Production, metabolic, and environmental data.	Artificial Neural Networks (ANN)	Production and reproduction traits, and environmental impacts, Prediction	(117–126)	(54, 127–135)	(102, 136–140)
Visual/time-series and sensor data.	Deep learning models	Visual identification, status monitoring, and early detection.	(141–149)	(150–158)	(116, 159–165)
Multi-sensor data, omics datasets.	Unsupervised learning	Animal phenotyping, anomaly detection,	(166–171)	(154, 172–177)	(178–184)