Hello,

We noticed you're browsing in private or incognito mode.

To continue reading this article, please exit incognito mode or log in.

Not an Insider? Subscribe now for unlimited access to online articles.

Business Report

Watson Makes a Salad

IBM’s Watson technology has digested years of research on flavor and taste and memorized 10,000 recipes. Now it’s creating recipes that its creators hope are both pleasant and surprising.

If it sounds like a salad created by a computer, that’s because it is. “Morel Mushrooms Leek Katsuobushi Hazelnut Pear Poached” is the recipe that Florian Pinel, lead engineer at the IBM Watson Group, and I came up with (well, mostly Pinel did). To make it we used a beta version of an online extension of the new cookbook Cognitive Cooking with Chef Watson. This online engine, powered by IBM’s Jeopardy!-beating cognitive computing system Watson, allows you to “discover” a recipe by selecting a few key ingredients and then exploring other foods with similar “flavor compounds” in a matchmaking exercise similar to Pandora’s method of grouping musical acts by common traits. You can choose to pair the initially chosen foods with those it identifies or substitute one for another.

Pinel, a chef as well as a researcher, had the idea of pairing a spring classic, the morel mushroom, with Japanese bonito fish flakes (katsuobushi), two foods that share a surprising number of common flavors—120 altogether, according to Watson. Using a Western approach—Westerners like recipes consisting of ingredients with shared flavor compounds, while Eastern cuisines pair opposites—Watson then suggested other ingredients that it predicted would taste good with morels and fish flakes, including leek, hazelnut, and pear. Watson then searched its memory banks of the 10,000 recipes published by Bon Appétit magazine over the past decades to suggest a suitable preparation, a twist on a classic poached pear and blue cheese salad.

This story is part of our July/August 2015 Issue
See the rest of the issue
Subscribe

To test Watson’s culinary imagination, I went shopping, printout in hand. I had trouble finding morels—springtime comes late to Boston—but the fish flakes were easily located in the international aisle at Whole Foods. For the white wine in the dressing I chose a white Bordeaux. In the kitchen, I found Watson’s cooking directions to be a bit sparse and sometimes odd (1¼ heads of lettuce, for example), so I did some independent research: I consulted ­Martha Stewart for how long it takes to poach a pear (20 minutes) and Epicurious for directions on toasting hazelnuts. But overall, the recipe came together nicely, and the panel of taste testers (my family) liked the combination of crunchy nuts and sweet pear. The katsuobushi’s flavor was subtle, and I would recommend going light on the Parmesan, but overall I’d give a strong review to a chef who’ll never have the pleasure of tasting one of his own creations.

Become an MIT Technology Review Insider for in-depth analysis and unparalleled perspective.

Subscribe today
Next in this Business Report
High-Tech Food Chain

Data-driven, robotic and mobile technologies are transforming how we grow, process, distribute and buy our food.

Want more award-winning journalism? Subscribe to Insider Plus.
  • Insider Plus {! insider.prices.plus !}*

    {! insider.display.menuOptionsLabel !}

    Everything included in Insider Basic, plus the digital magazine, extensive archive, ad-free web experience, and discounts to partner offerings and MIT Technology Review events.

    See details+

    Print + Digital Magazine (6 bi-monthly issues)

    Unlimited online access including all articles, multimedia, and more

    The Download newsletter with top tech stories delivered daily to your inbox

    Technology Review PDF magazine archive, including articles, images, and covers dating back to 1899

    10% Discount to MIT Technology Review events and MIT Press

    Ad-free website experience

/3
You've read of three free articles this month. for unlimited online access. You've read of three free articles this month. for unlimited online access. This is your last free article this month. for unlimited online access. You've read all your free articles this month. for unlimited online access. You've read of three free articles this month. for more, or for unlimited online access. for two more free articles, or for unlimited online access.