Think of how many times that you've had trouble picking a restaurant. It's incredibly hard actually deciding where to go, especially when you're in a large group, and everyone "doesn't care" where. Or perhaps you ask your significant other where to eat, yet you are both too "polite" to actually make a decision, and would both rather defer the question to the other.

What it does

foodme asks the user 4 simple questions then decides based on those and previous answers where the user should go. It then gives the user links to the Yelp page for the restaurant for more info, and a link to Google Map directions. At the end of the experience, the user can like or dislike the restaurant, affecting future decisions the app makes.

How we built it

The app was coded in Objective-C using Xcode, and relies on the Yelp and Google Maps APIs.

Challenges we ran into

  • Creating an effective machine learning algorithm and training it properly before release.
  • Designing a set of questions that minimizes user input, but still gets good restaurant results
  • Keeping the UI clean and visually appealing

Accomplishments that we're proud of

  • Our machine learning algorithm for this project
  • Completing an app in a weekend

What we learned

  • Two of our members learned iOS development for their first actual app
  • Using simple HTTP APIs and parsing through HTTP responses
  • Simple machine learning algorithms.

What's next for foodme

  • Facebook integration - suggest restaurants for your friends
  • Built-in navigation system - to allow the user to be completely surprised by the restaurant once they get there
  • Android version
Share this project: