Ajay Nathan, Alexander Suh, Gabriela Merz
Just as Netflix makes recommendations about movies you would like to watch, Prophit makes recommendations to Capital One customers about stores they would like to try. Since Capital One has data on their customers' financial situation and transaction history, they are in a unique position to make recommendations about future purchases, especially when that data on individuals is combined with the information Capital One has on its merchants, as well as Wolfram Alpha's macro data on the state of the economy. Prophit combines all of these datapoints in a machine-learning algorithm to predict the kinds of stores that customers will like, allowing financial institutions like Capital One to help their customers in a new way.
Make sure you are using a virtual environment of some sort (e.g. virtualenv or
pyenv), and make sure you have python 3.
pip3 install -r requirements.txt
./manage.py migrate
./manage.py loaddata sites
./manage.py runserver
Navigate to the capitalOne folder containing manage.py and do
python3 manage.py runserver
Then open up a browser, go to localhost:8000, and you're at the site! Since we made this site under the premise that only Capital One users could use it, there is no sign-up form on the site: hypothetically, users would sign up by signing up for an account with Capital One. However, so that you can experience the site, we've made an account for you! The username is 'capital', and the password is 'one'. Enjoy!
This project won "Best Use of Capital One's API" at HackHarvard 2015