Now showing 1 - 3 of 3
  • Publication
    Bayesian Case-Exclusion and Explainable AI (XAI) for Sustainable Farming
    Smart agriculture (SmartAg) has emerged as a rich domain for AI-driven decision support systems (DSS); however, it is often challenged by user-adoption issues. This paper reports a case-based reasoning system, PBI-CBR, that predicts grass growth for dairy farmers, that combines predictive accuracy and explanations to improve user adoption. PBI-CBR’s key novelty is its use of Bayesian methods for case-base maintenance in a regression domain. Experiments report the tradeoff between predictive accuracy and explanatory capability for different variants of PBI-CBR, and how updating Bayesian priors each year improves performance.
  • Publication
    Associations between paratuberculosis ELISA results and test-day records of cows enrolled in the Irish Johne's Disease Control Program
    The effect of the Mycobacterium avium ssp. paratuberculosis (MAP) ELISA status on test-day milk performance of cows from Irish herds enrolled in the pilot national voluntary Johne's disease control program during 2013 to 2015 was estimated. A data set comprising 92,854 cows and 592,623 complete test-day records distributed across 1,700 herds was used in this study. The resulting ELISA outcome (negative, inconclusive, and positive) of each cow within each year of the program was used to allocate the cow into different scenarios representing the MAP status. At MAPscenario1, all cows testing ELISA nonnegative (i.e., inconclusive and positive) were assigned a MAP-positive status; at MAPscenario2 only cows testing ELISA-positive were assigned a MAP-positive status; at MAPscenario3 only cows testing ELISA nonnegative (inconclusive or positive) and gathered exclusively from herds where at least 2 further ELISA nonnegative (inconclusive or positive) cows were found were assigned a MAP-positive status; at MAPscenario4 only cows testing ELISA-positive that were gathered exclusively from herds where at least 2 further ELISA-positive cows were found were assigned a MAP-positive status. Milk outputs based on test-day records were standardized for fat and protein contents (SMY) and the effect of MAP ELISA status on the SMY was estimated by a linear mixed effects model structure. The SMY mean difference recorded at test day between cows with a MAP-positive status and those with a MAP-negative status within MAPscenario1 was estimated at -0.182 kg/test day; the mean difference was -0.297 kg/test day for MAPscenario2; for MAPscenario3 mean difference between MAP-positive status and MAP test-negative cows was -0.209 kg/test day, and for MAPscenario4, the difference was -0.326 kg/test day.
      325Scopus© Citations 4