Cook up a Surprise with Artificial Intelligence-enabled Chef Watson

The Artificial Intelligence-human collaboration enables human chefs to explore creative culinary ideas in their own kitchens

What’s the link between Artificial Intelligence and Cooking? Nothing, they are as different as chalk and cheese, you may argue. Agreed. But, here comes IBM’s cognitive computing application—Chef Watson—that’s all set to revolutionize the culinary world.

Combinations and Permutations

Chef Watson helps cooks around the world to make innovative use of ingredients and make creative dishes. According to IBM, Chef Watson has analyzed about 10,000 recipes from the Bon Appetit application and knows the chemical composition of thousands of ingredients. Based on this data and the patterns, Chef Watson can suggest up to four ingredients that can be creatively used together. On the face of it, four may appear to be an insignificant contribution. However, when you consider suggestions for thousands of ingredients, you have a whopping quintillion combinations!

How Does it Work?

This is how it works. Once a human chef has selected a dish (cuisine) or an ingredient, Chef Watson uses its algorithms to suggest four ingredients that blend well with the selection. The human chef can select or discard the suggested ingredients. Chef Watson then explores its database and based on the chosen ingredients suggests hundreds of recipes from Bon Appetit. The human chef gets a detailed list of ingredients and instructions to the recipe. However, the human chef can make her own changes to the recipe and customize the dish by tweaking the ingredients. The collaboration enables human chefs to explore combinations that they never thought possible and discover new ideas in their own kitchens.

Artificial Intelligence at the Core

At the heart of the project is cognitive computing that enables applications to comprehend information much like humans do. These applications use natural language processing to understand the meaning of the words or sentences in the huge amounts of data used as training sets. They can also unravel hidden patterns from volumes of data and keep learning from the feedback to become smarter. A lot of research is underway to explore versions of Watson that can help analyze data on healthcare and diseases such as cancer to explore patterns that can help treat such diseases.

The article has been written by Neetu Katyal, Content and Marketing Consultant

She can be reached here.

Leave a Reply

Your email address will not be published. Required fields are marked *