I’m addicted to “discovery apps” — software that learns what you love, then recommends new options based on your preferences.
So I was delighted to come across Alfred, an app from CleverSense. Alfred learns about the restaurants you like in your neighborhood: where you like to go for breakfast, for brunch, for lunch, for dinner, for coffee and dessert. Then, based on those preferences, Alfred recommends nearby restaurants wherever you are: at home or on the road.
Especially when traveling, I use UrbanSpoon and TripAdvisor to discover great restaurants all the time. But both neither UrbanSpoon nor TripAdvisor actually make recommendations based on *my* preferences; instead, they rank eateries based on the preferences of everyone who bothers to vote.
And while the wisdom of The Crowd can be useful — especially if you have nothing else to go on — it can also be flawed. In the suburbs, for example, The Crowd is apt to rank McDonald’s as a great diner or Pizza Hut as fine Italian restaurant. And that’s ultimately the problem with ranking apps: systems that average together the votes of average diners tend to point to average results.
Alfred is different, in that his recommendations are, at least in theory, based on your preferences — the specific experiences you *personally* enjoy. I spent a little time training him (telling him about restaurants we go to on a regular basis), and in minutes, Alfred was recommending places within a mile or two of our house I’d either forgotten about (Tin Lizzy), avoided (Mojito), or never even heard of (Cafe 640?!?).
It’s easy to get excited about recommendations before they’re road-tested, so stay tuned. Meanwhile, over the next couple of weeks, I’m going to depend on Alfred to answer “What’s for dinner?” If you do the same, let me know how things go.
Add comment