Our final writing was done by Abby McCalmon. She is a second year master’s student in the Department of Animal Science at the University of Tennessee under Dr. Elizabeth Eckelkamp. She holds a B.S. in Animal Science with a minor in Biology from the same institution. She works with Precision Technology in the Dairy Industry especially Automatic Milking Systems and fetch cow status.
The ARPAS Symposium: Artificial Intelligence and Machine Learning in Dairy Production Systems
The ARPAS Symposium: Artificial Intelligence and Machine Learning in Dairy Production Systems was conducted Tuesday at the ADSA Annual Meeting. This hybrid forum pertained to opportunities for Precision Livestock Farming (PFL) in the dairy industry.
Dr. Joao Doreo from the University of Wisconsin-Madison started the symposium by defining AI technology and how it affects the world today. The objective of the talk was to show how AI has the potential to revolutionize the entire livestock industry. Some examples include computer vision systems, speech to text and mixed reality.
The second speaker was Dr. Albert De Vries from the University of Florida, who talked about examples and opportunities for artificial intelligence on the dairy farm. The goal of this talk was to give examples of AI on dairy farms and focus on some emerging opportunities. He mentioned the challenges of data collection from these systems but continued to explain solving this would advance the dairy industry far more than we could imagine.
Dr. Catie McVey from the University of California – Davis was our third speaker presenting on Livestock Informatics toolkit with visualizing complex behavior patterns across multiple PLF sensors using unsupervised machine learning. She introduced the LIT toolkit as a new open source code package through R that visualized complex nonlinear patterns in livestock data streams. This model showed evidence of a competing behavior mechanism linking time budget and entry order that would be overlooked in a linear model.
The final speaker was Dr. Juan Steibel, of Michigan State University, who talked on computer vision and machine learning for phenotyping. Steibel went over two real life examples of novel phenotyping using data collected on pig production systems: (1) the prediction of interactions at the feeder using video recording and (2) the use of statistical and machine learning methods to predict animal interactions using genomics data.
This session informed viewers on past, present, and future research regarding AI systems as well as vocalized the challenges that AI overcomes and the challenges it may cause.