Some of us work closely with political events and processes daily, so we felt it's important to analyze publicly available sources of political debate to understand what is really happening, why, and how to avoid an era where the biggest liar always wins. ;)
What it does
Our app shows the user quotes. The user needs to guess, if the quote is real or fake. Afterwards the user is given feedback on his input. We also analyse user's micro-movements to rate his or her confidence level.
How we built it
We used React Native to build the mobile game app, Node.js and PostgresQL for the backend, and Python and TensorFlow to analyze the Politifact.com data
Challenges we ran into
- Inconsistent duplicates of statements while scraping 12,500 statements
- A missing comma
- AWS sucked. :(
Accomplishments that we're proud of
- Amazing logo
- Innovative algorithm that takes Tinder-like gestures and analyze.
- Amazing backend
What we learned
- Learned to use pandas, seaborn, jupyter notebooks, TensorFlow, etc. for the first time.
- Learned that knowing how to set-up server infrastructure has become irrelevant in 2016.
- This is 200% true.
What's next for Awful Press
Oh, boy! We're gonna be on the first page of The Sun.... Wait.... No, we're not. :D
Log in or sign up for Devpost to join the conversation.