All posts filed under: Big Data

Big data supports expert wine tasters

In the course of developing software for predicting consumer wine preferences, a Houston-based start up, VineSleuth, shed new light on the abilities of expert wine tasters and the validity of blind tasting assessments. Contrary to popular belief, the company’s VineSleuth metrics, which are based on the work of Chief Science Officer Michael Tompkins and his team, reveal that tasters can consistently identify aroma and flavor characteristics in blind wine evaluations. “We have extensive experimental data which support that expert evaluators have the capacity to precisely identify wine characteristics in blind repeat samples,” said Tompkins whose work spans thirteen years in the field of numerical methods. “During the course of our experiments, our vetted evaluators repeat sample characteristics about 90% of the time,” he says. VineSleuth’s data directly confronts the popular misconception that consistent sensory evaluation of wine is a random occurrence. In developing an algorithm designed to help consumers make wine selections based on personal preference, the company has established a benchmark based on the results of its top-performing tasters (including this author) and intends …