It's lunchtime, you are looking for somewhere to go to eat so you open Yelp and look for recommendations. After scrolling through many pages, you are overwhelmed by the number of restaurants around you and can't decide where to eat so you end up going to the fast food restaurant you always go to. We've all been there. What others like may not be what you like, but you also do not want to waste time entering all your preferences on the app. But wait, you always like photos on social media, so shouldn't your phone know what you like already?
What it does
Doko will collect data about the photos foods/restaurants the user have liked on social media and next time the user pass by that location, we will notify you. The user can also see restaurants around him/her in a convenient map view. Since t you have shown interest already in these restaurants, we are confident in our recommendations.
How we built it
We used Twitter API to query tweets a certain user has liked every 10 seconds. The backend is written in Python serving as the API connecting Mongo DB and iOS.
Challenges we ran into
Getting trapped by MongoDB Stitch iOS SDK. It took us nearly 2 hours to find out the issue in our project (also the inexplicit of the document) after reading the source code of it.
Accomplishments that we're proud of
What we learned
What's next for Doko
Our first step will be to support social media other than Twitter. Then we can include additional features such as restaurant recommendations using machine learning algorithms, or making a reservation within the app.