Semantic technology can be part of the fun. Over the next couple of days we’ll look at some ways it can chip in. Let’s start with food as you start thinking about the summer BBQs. There are semantic solutions that can help on various fronts here. Edamam, for example, has built a food ontology that classifies ingredients, nutrients and food that it applies to recipes it scrapes from the web with the help of its natural language processing and machine learning functions.
As you’re breaking out the grill, you can break out the smartphone or iPad to search for grilled burger recipes that incorporate tomatoes in the 200 to 400 calorie range, for example, and take your pick of ranch salmon, Portobello mushroom, turkey with spiced tomato chutney or the classic beef with garden vegetables, for instance. “The nutrition information we append to recipes using natural language processing. This translates into people being able to filter recipes by diet/calories/allergies and be a bit more health-conscious this summer,” says Victor Penev, Edamam founder and CEO.
“We also have a nutrition wizard, which allows people to analyze the food they cook and get a sense of the nutrients (fat, calories, protein) they get. It’s another way to keep yourself a bit more healthy,” he says. Epicurious users who add their own recipes to its member recipe database, for example, can get the inside scoop on just what they’re consuming from the family’s food treasure trove, as well as from most of the site’s 400,000 recipes.
Yummly offers semantic-flavored recipe search, too, as well as a Recipe API that lets developers integrate recipes and faceted recipe search into websites and mobile apps. Its spring update (which we covered here) saw its semantic recipe search and filtering — by categories including occasion and holiday — across more than a million recipes make its way to the iPad, too, for grillside advice.
Over at Klappo, which uses semantics to add additional intelligence to food data, the Sensum.io commercial API built atop the platform’s semantic knowledge base of nutritional data for packaged products and recipes also lets app developers and food suppliers build services and mobile apps to guide their customers about food choices, whether they need to avoid certain ingredients for health issues, including interference with prescription drugs, or to maintain adherence to religious principles. That’s in beta right now, but the Klappo technology platform itself claims to include more than 60,000 ingredients with nutritional information from 30 countries; 130,000 packaged products with nutritional data and information; and a half-million recipes parsed and annotated with full lists of ingredients, processing and nutritional info. Its NLP technology, the company says, allows realtime mapping of recipes, from ingredient extraction to prepared dish, including mapping that information against intolerances and allergies. (Hey, no one wants Aunt Mary to get sick at the big family barbeque!) And its semantic smarts paves the way for it to understand food down to the level of the effects of combining ingredients.
There are other solutions that help inform recipe searches, as well, such as the U.K.-oriented Whisk (though I’m not sure how well the weather there lends itself to regular outdoor summer living).
But enough about how to find out more about what’s a good outdoor drink or meal. What about finding out what your best grill option is? That’s where work like what Best Buy has undertaken with building semantic product and review endpoints — and whose APIs for products, stores, reviews, categories and more are being leveraged in its own Best Buy App, as well as third parties like TrackIf — can come in handy. With the Best Buy App, for example, the structured data that infuses the vendor’s efforts enables users to click-search their way through grills (and other items) via price range, customer reviews, special offers, free shipping deals, and more.
Walmart’s in the mix too, when it comes to finding the perfect grill (or anything else) with its Polaris search engine that grew out of the semantic technology it got via its Kosmix acquisition, and to whose features its recent acquisition of e-commerce technology company Adchemy and its talent is expected to further contribute. The Kosmix built internal search engine led to a 20 percent increase in search conversion, Walmart reports. In fact, Walmart’s slate of recent acquisitions also lead us back to what’s cooking on that grill: In February it acquired Yumprint, which understands recipe semantics and consumer taste preferences to help users discover recipes, matches ingredients to advertisements, calculates nutritional information and prepares shopping lists from recipes.
Walmart in a blog at the time of the deal said Yumprint’s founders “ideas and ambitions for transforming the grocery shopping experience match the global opportunity Walmart enjoys in this space, and their accomplishments with Yumprint just scratch the surface of what we’re going to do next together.”