Translating Big Data into Smart Data for Veterinary Epidemiology

VanderWaal, Kimberly; Morrison, Robert B.; Neuhauser, Claudia; Vilalta, Carles; Perez, Andres M.

The advent of big data has implications for the education of veterinary epidemiologists (6, 13, 17, 51) , including technical skills, such as computer programming, that may not be a traditional part of epidemiological training. While epidemiologists may never be responsible for creating complete software applications, the ability to manage relational databases or write simple scripts in a programming language to facilitate preparing data for analysis is critical when datasets become too large to process manually. Further, the analysis of big data often entails the use of supercomputing resources, which usually requires some familiarity with parallel processing and IT systems. To train the current workforce, workshops with hands-on computational activities are needed. Current curricula in graduate education should be expanded to include machine learning as well as traditional statistics, and coding as well as core epidemiological skills. Alternatively, graduate programs in the veterinary sciences could actively recruit students with computer science backgrounds that will readily be able to apply big data thinking to veterinary data. Veterinary epidemiologists with skillsets that allow them to directly engage with, manipulate, and analyze large datasets will be ideally situated to propel veterinary epidemiological research and practice into the coming decades.