Flavor Pairing in Medieval European Cuisine: A Study in Cooking with Dirty Data

摘要

An important part of cooking with computers is using statistical methods tocreate new, flavorful ingredient combinations. The flavor pairing hypothesisstates that culinary ingredients with common chemical flavor components combinewell to produce pleasant dishes. It has been recently shown that this designprinciple is a basis for modern Western cuisine and is reversed for Asiancuisine. Such data-driven analysis compares the chemistry of ingredients to ingredientsets found in recipes. However, analytics-based generation of novel flavorprofiles can only be as good as the underlying chemical and recipe data.Incomplete, inaccurate, and irrelevant data may degrade flavor pairinginferences. Chemical data on flavor compounds is incomplete due to the natureof the experiments that must be conducted to obtain it. Recipe data may haveissues due to text parsing errors, imprecision in textual descriptions ofingredients, and the fact that the same ingredient may be known by differentnames in different recipes. Moreover, the process of matching ingredients inchemical data and recipe data may be fraught with mistakes. Much of the`dirtiness' of the data cannot be cleansed even with manual curation. In this work, we collect a new data set of recipes from Medieval Europebefore the Columbian Exchange and investigate the flavor pairing hypothesishistorically. To investigate the role of data incompleteness and error as partof this hypothesis testing, we use two separate chemical compound data setswith different levels of cleanliness. Notably, the different data sets giveconflicting conclusions about the flavor pairing hypothesis in Medieval Europe.As a contribution towards social science, we obtain inferences about theevolution of culinary arts when many new ingredients are suddenly madeavailable.