-
T(r/h)um(p/bs) up!
Inspiration
Trump's tweets
What it does
It takes a real article as user input, infers most important words from the article, and finally generates text primed with those most important keywords.
How we built it
We used python based machine learning as well as python web frameworks to build the thing.
Challenges we ran into
Our neural networks did not learn or do anything.... Also had some minor issues with web scraping and data mining as well. Google seems to not like this.
Accomplishments that we're proud of
It really feels like a fake fake Trump!
What we learned
Data science, team work, connecting machine learning workflows into a web app.
What's next for Trumpulator
Conquer the world.
Log in or sign up for Devpost to join the conversation.