A Stacking-Based Model for Animal Disease Detection in Burkina Faso
- Lecture Notes in Networks and Systems , 1893 : 390-399
Résumé
Livestock farming in Burkina Faso faces numerous challenges, such as the reduction and degradation of grazing areas, difficulties in accessing pastoral and land resources, as well as the prevention and management of diseases, particularly epizootics. Although the number of qualified agents appears sufficient to address some of these issues, the field of animal health suffers from a limited number of veterinarians and para veterinarians, making it difficult to effectively manage both public and animal health. In this paper, we propose an approach based on Artificial Intelligence (AI), specifically based on the ensemble stacking method, for the diagnosis of animal diseases. Our process begins with data collection to build a medical diagnostic model. During model construction, several machine learning algorithms were tested, and the top three (03) were combined using the stacking method to create a final meta-model achieving a remarkable accuracy rate of 99.01%. This research thus highlights the great potential of AI technologies in the field of animal disease diagnosis, offering an effective and promising approach to improve animal health and veterinary practices in Burkina Faso.
Mots-clés
animal disease diagnosis, stacking ensemble, meta-model, livestock health, epizootic, veterinary practice