Title: Food Pairing Unveiled: Exploring Recipe Creation Dynamics through Recommender Systems

Authors: Giovanni Palermo (Physics Department, University ”Sapienza”, Enrico Fermi Research Centre – Roma), Claudio Caprioli (Department of Physics and Astronomy ”Ettore Majorana”- Catania), Giambattista Albora (Enrico Fermi Research Centre – Roma)

Published in: Archive Physics June 21 2024

Link: arXiv:2406.15533

Abstract:

In the early 2000s, renowned chef Heston Blumenthal formulated his “food pairing” hypothesis, positing that if foods share many flavour compounds, they tend to taste good when eaten together. In 2011, Ahn et al. conducted a study using a dataset of recipes, ingredients, and flavour compounds, finding that, in Western cuisine, ingredients in recipes often share more flavour compounds than expected by chance, indicating a natural tendency towards food pairing. Building upon Ahn’s research, our work applies state-of-the-art collaborative filtering techniques to the dataset, providing a tool to recommend new foods to add to recipes, retrieve missing ingredients and advise against certain combinations. We create our recommender in two ways: by considering ingredients’ appearances in recipes or by sharing flavour compounds between foods. While our analysis confirms the existence of food pairing, the recipe-based recommender performs significantly better than the flavour-based one, concluding that food pairing is just one of the principles to consider when creating recipes. Furthermore, and more interestingly, we find that food pairing in data is mostly due to trivial couplings of similar ingredients, leading to a reconsideration of its current role in recipes, from being an already existing feature to a key to opening up new scenarios in gastronomy. Our flavour-based recommender can thus leverage this novel concept and provide a new tool to lead culinary innovation.