Inspiration
We wanted to find out whether the power of an influencer on social media like Twitter has tangible effects on short-term financial markets.
What it does
Thanks to the availability of Twitter's API, we were able to throw together a proof of concept which uses Donald Trump's prolific Twitter account to test if such an influencer has the ability to create fluctuations in stock prices.
How we built it
We first used a twitter scraper to collect the tweets from Trump's twitter as well as a stock scraper to get data off yahoo finance on various blue chip stocks. We then used pandas to coalesce the data and put it into a format easy for training and testing in a neural net. Finally, we constructed an ANN in Keras to predict stock market changes based on various measurable descriptors of Trump's tweets such as the number of the capital letters, the punctuation, and more.
Challenges we ran into
Time.
Accomplishments that we're proud of
Coalescing data from disparate sources together into something that can actually create a meaningful impact on the economy and society. The more Trump's staff understands how his tweets affect the markets, they can better manage the child-in-chief.
What we learned
how to use keras; finding value in two different data sets
What's next for SaveTheMarket
Integrating natural language processing and connecting front and back end so users can see how Trump's words themselves (i.e. tax cuts, tremendous, wall, #MAGA) impact the markets.
Log in or sign up for Devpost to join the conversation.