But can it guarantee that your food will taste as good as it looks on your phone?
Are you ready to try another recommender app? Even if you’ve been burned by less-than-stellar automatic recommendations in the past, free restaurant recommender app Ness is worth a shot. The iOS app, from Silicon Valley startup Ness Computing, boasts serious technology put together by an impressive team of engineers, all in the service of giving you tips on where to grab a bite.
Ness learns over time and builds a “taste profile” of you. The more it learns, the better the recommendations, a la Pandora. Facebook and Foursquare recommendations and check-ins from friends all show up on the Ness search results page.
The app makes recommendations by crunching together your taste profile, your similarity to other people, recommendations from friends, and overall popularity ratings.
The company’s Likeness Engine gives you a Likeness Score from 0 to 100 for each restaurant. I’m glad they’re producing fine-grained ratings. Five stars, even with half stars, have never felt adequate.
You can also filter the results. Choose “Hide Places I’ve Rated” to make sure new restaurants show up at the top of your search results. And choose “No Big Chains” to stick to local businesses. Of course, if you and your friends are foodies, the chains shouldn’t show up in your search results anyway, right?
Ness also lets you choose which of your friends influence the app. This is ostensibly to help you pare down extensive social networks, but I would use this feature to filter out my meat-and-potatoes-only friends. No matter how clever systems like this become, the Achilles heel is the quality of the data they’re drawing on–garbage in, garbage out.
Under the hood, Ness’ Likeness Engine uses the same sophisticated technology the spooks in Washington use to track terrorist networks: machine learning, collaborative filtering, social graph data mining and natural language processing.
Ness Computing isn’t stopping at food. They plan to expand the app to all kinds of categories where you want to know how much you like something before springing for it, including music, shopping, nightlife, and entertainment.
Now if it can tell me what I’m in the mood for…
(Via Giga Om)