Meet Sony’s New Flavor Crop and have researched over a million recipes to find out which ingredients work with which ingredients
If you’ve been running your TV in prime time for the past few weeks, it looks like a cooking show or cooking contest airs for an hour every day, in which we see chefs trying to combine ingredients and ingredients. Very true, we are not all born with the intuition “clothes and tears” that Sim Cohen wants to praise the most, but Sony’s new study may reveal the real reasons why some products work together, for which he certainly uses artificial intelligence. Overall, 2021 now.
More than a million recipes and more than 1,500 flavor molecules were donated to the organization
Sony’s AI department has launched a joint study with the University of Korea as part of its flagship gastronomy program launched in late 2020. At this time, it is releasing a new tool called Flavor Graph, which will eventually predict the results of combining two ingredients or foods. But we’re not just talking about fries and ketchup or clear combinations like peanut butter and jam, we’re not talking about obvious lesser ones like caviar and chocolate.
The researchers entered more than a million recipes on the computer (sorry, not sure if this will be the last one), but were not satisfied with that, and entered information about 1,561 flavor molecules. In this way they can analyze the chemical and statistical relationship between different products and successfully create different taste profiles and explain the relationships between them.
The first diagram published in the study was used for sample training only, and includes only 160 of the 6,653 different components analyzed by the computer. Researchers say that after the creation of the Flavor Crop they were able to predict the components that would work with each other, but, as expected, point out the limitations of the study, and claim that some of the matches found in it are not yet scientifically tested, so there is no clear way to explain. Do research in this space.
After creating the system, the researchers tried to get recommendations for foods and ingredients that work well between white wine and Results You’ll find fillets of the same, bass and capelin pasta, but you’ll also find slightly harder results like concentrated mushroom soup. So far, researchers have not yet released unexpected matchmaking that will cause you to fall off a chair and run into the kitchen, but as research becomes more complex and science gets clearer representations of foods, it may even find bizarre matching and alternatives to unhealthy or unhealthy ingredients.