arXiv subjects

Claudio Caprioli

Publications and source records attributed to Claudio Caprioli.

3 recordsLinked to original sources

arXiv2024

The networks of ingredient combinations as culinary fingerprints of world cuisines

Investigating how different ingredients are combined in popular dishes is crucial to uncover the principles behind food preferences. Here, we use data from public food repositories and network analysis to characterize and compare worldwide cuisines. Ingredients are first grouped into broader types, and each cuisine is then represented as a network in which nodes correspond to ingredient types and weighted links describe how frequently pairs of types co-occur in recipes. Cuisines differ not only in the popularity of ingredient types and range of recipe sizes, but also in the structural organization of ingredient-type combinations. By analyzing these networks, we uncover distinctive patterns of type associations that serve as culinary fingerprints. For example, European cuisines typically distribute ingredients across different types, whereas certain Asian and South American traditions emphasize one dominant type, such as vegetables or spices. The essence of these patterns is well captured by the networks' maximum spanning trees, which offer a simplified yet representative backbone for each cuisine. We demonstrate that both these full and simplified network representations enable machine learning models to identify cuisines from subsets of recipes with very high accuracy. Networks of ingredient combinations also cluster global cuisines into meaningful geo-cultural groups, reflecting shared patterns in culinary traditions. More broadly, our study offers novel insights into the structure of world cuisines, enabling data-driven approaches to their characterization, cross-cultural comparison, and potential adaptation.

physics.soc-ph↗

arXiv2024

Food Pairing Unveiled: Exploring Recipe Creation Dynamics through Recommender Systems

In the early 2000s, renowned chef Heston Blumenthal formulated his "food pairing" hypothesis, positing that if foods share many flavor compounds, then they tend to taste good when eaten together. In 2011, Ahn et al. conducted a study using a dataset of recipes, ingredients, and flavor compounds, finding that, in Western cuisine, ingredients in recipes often share more flavor 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 that can recommend new foods to add in recipes, retrieve missing ingredients and advise against certain combinations. We create our recommender in two ways, by taking into account ingredients appearances in recipes or shared flavor compounds between foods. While our analysis confirms the existence of food pairing, the recipe-based recommender performs significantly better than the flavor-based one, leading to the conclusion that food pairing is just one of the principles to take into account when creating recipes. Furthermore, and more interestingly, we find that food pairing in data is mostly due to trivial couplings of very similar ingredients, leading to a reconsideration of its current role in recipes, from being an already existing feature to a key to open up new scenarios in gastronomy. Our flavor-based recommender can thus leverage this novel concept and provide a new tool to lead culinary innovation.

physics.soc-ph↗

arXiv2022

Deep Neural Networks as Complex Networks

Deep Neural Networks are, from a physical perspective, graphs whose `links` and `vertices` iteratively process data and solve tasks sub-optimally. We use Complex Network Theory (CNT) to represents Deep Neural Networks (DNNs) as directed weighted graphs: within this framework, we introduce metrics to study DNNs as dynamical systems, with a granularity that spans from weights to layers, including neurons. CNT discriminates networks that differ in the number of parameters and neurons, the type of hidden layers and activations, and the objective task. We further show that our metrics discriminate low vs. high performing networks. CNT is a comprehensive method to reason about DNNs and a complementary approach to explain a model's behavior that is physically grounded to networks theory and goes beyond the well-studied input-output relation.

cs.LG↗