How IBM’s Watson Will Make Your Meals Tastier
Coleslaw is a fairly straightforward dish to make. And yet, Dawn Perry, Bon Appetit’s senior food editor, recently found herself staring at list of ingredients for a Fourth of July slaw that included a strange array of flavors: cabbage, tamarind, mayo, buttermilk, basil, two types of onions and flour. If you know anything about the summer side, this recipe was not normal.
“I was like, what heck am I going to do with flour in a coleslaw,” she recalls. “Plus I have all these onions to deal with.”
The odd culinary directive came from Watson, IBM’s cognitive computer best known for kicking Ken Jennings’ butt on Jeopardy in 2011. Since its game show victory, Watson has been beefing up its cooking skills. You can now refer to him as Chef Watson.
This week, Bon Appetit and IBM are releasing the beta version of a new app called Chef Watson with Bon Appetit that will help home chefs think up new and inspiring ways to use ingredients. Think of Watson as an algorithmically inclined sous chef that gently suggests hundreds of flavor combinations that you’d probably never come up with on your own.
To do this, Watson crawled a database of 9,000 Bon Appetit recipes looking for insights and patterns about how ingredients pair together, what style of food it is and how each food is prepared in an actual dish. When the computer combines this information with its already robust understanding of food chemistry and hedonic psychophysics (the psychology of what people find pleasant and unpleasant), you get a very smart kitchen assistant.In the app, the user will input what ingredients she wants to use or avoid in her dish. She’ll note the style based on Bon Appetit’s tags (italian, asian, quick and easy?) and the kind of dish she’s looking to make (burrito or pasta?). From there Watson will come up with quintillions of possible combinations and present the top 100 in groups of 10 that range from the most common to the most experimental. “It gets back to, can a computer help you create, ideate and invent?” says Adam Rapoport, Editor in Chief of Bon Appetit. ”There are infinite flavor combos out there that you’ve never bothered to explore just because you never think to.”
This might sound familiar. Earlier this year Chef Watson partnered with the Institute of Culinary Education to create recipes for a food truck at SXSW. Human chefs were given a series of flavor and ingredient combinations created by Watson, which they then turned into a tasty meal. The new app takes this a step further, bringing the algorithmically-driven menu to home chefs by suggesting how much of ingredients should be used, and how they might be best prepared.
Taking a step back, the app is yet another example of the advancements we’re currently seeing in cognitive computing capabilities. Instead of simply regurgitating information gleaned from a database, Watson is now able to form deeper, contextualized learnings about the data it gathers. “The goal for us is to demonstrate what it is cognitive systems are capable of doing and how they help people discover new things,” says Steve Abrams, director of IBM’s Watson Group. “We’re using food as a vehicle for that.” Abrams points to Watson’s work in the medical field to how this deeper understanding of the data it gathers could not be best used. Watson already helps doctors find patterns and possible drug treatments, now it might give suggestions on how to implement those treatments, too.
For what it’s worth, the flour, onions and buttermilk weren’t such weird suggestions for the coleslaw recipe after all. Perry battered the onions in the buttermilk and flour, fried them up and crushed them over the coleslaw to add a bit of fatty crunch. Turns out, even computers love onion rings.
