FeedEfficiencyService: an architecture for thecomparison of data from multiple studies related todairy cattle feed efficiency indices.

• Autor(a) principal: LINHARES, H. M.
• Data de Publicação: 2021
• Outros Autores: BRAGA, R., ARBEX, W. A., CAMPOS, M. M., CAMPOS, F., DAVID, J. M. N., STROELE, V.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da EMBRAPA (Repository Open Access to Scientific Information from EMBRAPA - Alice)
• Texto Completo: http://www.alice.cnptia.embrapa.br/alice/handle/doc/1136046; https://doi.org/10.1016/j.inpa.2021.07.002
• topic: Gado Leiteiro; Alimento Animal; Alimentação; Eficiência Nutricional

The increased demand for food worldwide, the reduced land availability for livestock production, the increasing cost of animal feed and the need for mitigating livestock-related greenhouse gas emissions have driven the search for animal feeding systems that proves more efficient. To tackle this problem, we propose the use of computational support to help researchers compare data on feed efficiency, therefore improving economic and environmental gains. As a solution, we present an integrative architecture capable of combining heterogeneous data from multiple experiments related to dairy cattle feed efficiency indices. The proposed architecture, called FeedEfficiencyService, classifies animals according to feed efficiency indices and allows visualizations through ontologies and inference engines. The results obtained from a case study with researchers from the Brazilian Agricultural Research Corporation ? Dairy Cattle (EMBRAPA) demonstrate that this architecture is a supporting tool in their daily work routine. The researchers highlighted the importance of the proposed architecture as it allows analyzing animal data, comparing experiments, having reliable data analyses, and standardizing and organizing data from experiments. The novelty of our approach is the use of ontologies and inference engines to enable the discovery of new knowledge and new relationships between data from feed efficiencyrelated experiments. We store such data, relationships, and analyses of results in an integrated repository. This solution ensures unified access to the processing history and data from diverse experiments, including those conducted at external research centers.