-
-
Menu of Marciano(with like feature)
-
Home Page with the description of the three dining halls.
-
Menu of Warren(with like feature)
-
Menu of West(with like feature)
-
Recommendation page. Give personal recommendation for the user.
Track - Innovating with AI // Sponsored challenge - BU Spark: Best College Life Hack
Inspiration
There is no way to easily access the menus of Boston University's dining halls, and they have so many food options we sometimes don't even know which one to chow down!
What it does
It provides the menus of BU's dining halls all in one place, and it records the user's previous preferences (the user can choose to 'like' the food if they love it) to recommend food based on what they loved to eat before.
How we built it
We used Flask for the backend, Next.js for the frontend, and Machine Learning using python.
Challenges we ran into
We faced challenges trying to connect the frontend, backend and ML model together. However, after a long period of debugging, we managed to make it work!
Accomplishments that we're proud of
We are proud to be able to successfully design and use APIs to get data. We were able to render that data in really interesting ways on the frontend! The frontend was able to come to life with the data and results that our ML model produced!
What we learned
We learned new things about the tools we were using such as Flask and Next.js. Through many hours of struggle we discovered tricks to parsing data, styling CSS and making API calls.
What's next for BU Personal Dining
We look to making our code more efficient by caching our data and improving the API. We also envision using a better ML model called the k-th nearest neighbor algorithm to replace the similarity algorithm we have currently, which is not as precise. Once we have more users, we can also categorize the users according to their dietary preferences, such as vegetarian, vegan, kosher, and that will further improve our recommendation system.
Log in or sign up for Devpost to join the conversation.