En anglais seulement.
This project utilizes Large Language Models (LLMs) and cross-species transfer learning (chicken and salmon data) to predict broiler livability. By analyzing smart farm sensor data and production factors like temperature changes and FCR, the model achieves a high-reliability prediction (less than 2% error). The outcome provides farmers with AI-driven recommendations to optimize conditions and significantly reduce chick loss during the critical first ten days.
Présentateur :
Rezwan Ghaffarzadeh Derakhshan Faculté de médecine vétérinaire, Université de Montréal, 3200 Sicotte, St-Hyacinthe (QC) J2S 2M2 rezwan.ghaffarzadeh.derakhshan@umontreal.ca
Superviseur(e) principal(e) :
Auteurs, coauteurs : Rezwan Ghaffarzadeh Derakhshan, FMV, Universite de Montreal
Some text in the modal.