Inspiration
Ever been overwhelmed by the top 20 on Yelp? Spending more than 20 minutes just deciding on what to have for dinner? Not sure what to eat but don't want to regret your decisions? Munchii is here to help! It's a personalized food buddy that knows exactly what you want, adapting to whatever time/place/situation you are in.
What it does
Munchii is simple, and smart. It tells you where you should go, and what you should eat. Every time you give it feedback, it learns from your eating habits and makes better suggestions next time. The more you use Munchii, the more it knows about what's important to you, be it the type of cuisine, the location, or even the user reviews!
How we built it
Munchii is a web app built with Django, aimed to be a light weight application that can run on all types of devices. The front end is built responsively using javascript and jQuery to maximize user experience. The machine learning system in Munchii utilizes linear regression using gradient descent on the high dimension of features gathered from Yelp API, and enhanced by Indico's sentiment analysis plugin.
Challenges we ran into
We wasted the first 12 hours of the Hackathon on a different idea but it didn't materialize, so we built Munchii in the last 20 hours of HackTheNorth, working until the last minute. We have never used most of the libraries we used in this project, so we had to learn everything on the fly.
Accomplishments that we're proud of
Oksana is proud of being on TV. Venus is proud of pulling her first all nighter. Casper is proud of his snapstory.
What's next for Munchii
To polish the user experience, and add more personalization abilities to the application. Include multi-user support for group outings, quickly find out each person's preferences without ever having to debate over food again!
Log in or sign up for Devpost to join the conversation.