CollegeFoodie

Logo
College Foodie Landing Page
See Food Ratings and Recommendation
Rate Foods At Your College
Firebase Database
Log In to College Foodie
User Authentication

Inspiration

Rate My Professor allows college students to share information about the teachers around them, helping them make more informed decisions about courses. However, college students have another problem. They often struggle to find good and healthy food on their campuses, even when that food exists. Students need a platform to share their recommendations for the best food around their college campuses, so that they don't have to sit through mystery meat and overcooked pasta.

What it does

College Foodie is a website where users can create a profile, rate foods at their college, and get recommendations of meals liked by other similar users. The website is built around a recommender system, a predictive algorithm that can offer more accurate food recommendations when more users and ratings are put into the system.

How we built it

We constructed the front end of our web application using a combination of HTML, CSS, Bootstrap, and Materialize. We used Firebase Authentication and ReCaptcha to allow users to create accounts and log in. Then, for specific users, we used Firebase Database to store user data around food ratings. The web application then runs a recommender system on Javascript to predict (with increasing accuracy) what other foods a user might enjoy. This recommender system analyzes data with a user-to-user similarity matrix and the Pearson Correlation as our significant weighting. We also built our web application with a focus on security. Firebase is secure and anonymous. We cannot see user passwords, and ratings can only be added when a user is signed in. Additionally, ReCaptcha protects against bots and other attacks.

Challenges we ran into

As a team made almost entirely of first-time hackers, we struggled to limit our project to a scope that could be achieved within a single day. We also were learning technologies as we implemented them. On the front end, learning Bootstrap was a challenge. On the backend, Firebase was time-consuming. We were surprised that merging our work between frontend and backend was so challenging, especially with our work regarding ReCaptcha. Together, these things meant that we didn't get to implement some of the more complex aspects that we'd hoped to get to.

Accomplishments that we're proud of

We're proud of implementing a recommender system that can accurately learn user preferences and predict what foods they might like. We're also proud to have participated in our first Hackathon!

What we learned

We learned Bootstrap, Firebase, how time-consuming team web development can be, how challenging it can be to merge code, and how to do ReCaptcha.

What's next for CollegeFoodie?

We hope to add more filters (such as schools and dietary restrictions), fully integrate Recaptcha, and add photos to go with the foods on the web page.