FlavorGraph by Sony uses AI to pair ingredients with each other

After making its way of success into games, self-driving, and other areas, AI is now trying its hand at cooking, thanks to Sony AI and Korean University’s new ingredient mapping tool called FlavorGraph. This project is designed to meaningfully pair up ingredients.

As Sony and Korean University (KU) researchers noted in a blog post that the chefs mastered pairing ingredients through their intuition rather than through scientific knowledge or statistical learning. Resulting in a cultural evolution for many pairings like – cheese and tomato, pork and apple, and garlic and ginger. Many of those pairing were later explained by science after researchers realized that how ingredients sharing dominant flavor molecules paired well together.

“Ingredient pairing is an excellent challenge for AI systems. It requires AI systems to build models of how people pair ingredients from recipes as well as through other strategies. This work is a great step in AI for creativity within our Gastronomy project.”

The team consisting of Donghyeon Park, Keonwoo Kim, and Professor Jaewoo Kang examined both molecular information and how chefs and recipe creators paired the ingredients in their recipes. And then the team created the FlavorGraph database based on the data of 1,561 flavor molecules and of a million recipes.

The resulting data of FlavorGraph shows suggestions to achieve better results than ever before. And these can be used to predict relationships between compounds and foods by chemical compounds present in them and how they affect their overall taste. This will further help to hint at “new and exciting recipe techniques and driving new perspectives on food science in general.”

“As well as relationships between food ingredients and flavor compounds that have not been explored before, the FlavorGraph research will allow greater flexibility for matching single or multiple ingredients to many others.”

“As the science develops and we get ever better representations of food, we should discover more and more intriguing pairings of ingredients, as well as new substitutes for ingredients that are either unhealthy or unsustainable.” the team wrote.

FlavorGraph is just the starting point for AI in the cooking area, but in the near future projects like this “will help us to discover new recipes, changing the way we eat forever.”