The way foods and beverages taste—their set of flavors—is extremely important to most people. Scientists who develop and study food products must use tasting panels of human volunteers to evaluate the flavors of foods. These panels create popular tools—like the “wine wheel” used by countless wine lovers and professionals—and use them to make important processing and economic decisions. However, human panels are expensive, time-consuming, and limited in the number of products they can evaluate. The proliferation of digital review and discussion sites about food products—for example, RateBeer—has provided a rich new source of product descriptions. The field of Natural Language Processing (NLP) has developed to deal with this kind of abundant textual data, but has not been previously applied to descriptions of tastes and flavors. We think it is possible to use NLP techniques on digital reviews of food products in order to develop useful flavor tools without a human panel. The Seeing Flavors project aims to take advantage of more than 6500 online whiskey reviews—a dataset orders of magnitude larger than those usually available in food science—in order to develop practical and interactive “flavor wheels” for whiskeys. Anyone with any level of interest in whiskey will be able to use these tools to visualize the flavors of specific whiskeys. For example, we plan to develop an interactive tool that will visualize the relationships between whiskey categories—bourbon or scotch, young or old—and specific flavors. The same type of tool, we hope, could be used to develop a map of the world of whiskey, showing which whiskeys share similar flavors, and which flavors are related to overall whiskey quality, price, and other relevant attributes.
Looking into the future, we anticipate that the methods we will develop to accomplish these goals will be transferable to other products, and that the application of NLP tools to food science goals will allow for more rapid, thorough, and comprehensive flavor research. The hope is that by making flavors visible, the application of NLP to big datasets of food reviews will benefit both consumers and producers of a variety of foods and beverages.