The stonks team was inspired by the subreddit wallstreet bets(wsb) as recently it has influenced several market trends. We wanted to see if we could use this data to predict whether it was a good or great time to buy.
What it does
stonks utilizes a system of our creation that weights mentions and flairs in wallstreet bets with traditional financial analysis. We then display our findings to the stonks webpage
How we built it
We used flask and HTML to create our web app. Python was used in machine learning, analysis and connecting our front and back ends. SQl was used to set up databases to train our system.
Challenges we ran into
Implementing machine learning and coordinating API requests with csv files was a challenge as data had to be processed before we could run any real analysis on it.
Likewise, traditional financial analysis and trading nuances proved to be difficult both selecting on relevant data and analyzing it proved to be a challenge
Accomplishments that we're proud of
We are proud of our frontend we utilized different technologies than we would typically use. We are also proud of the work and research we put into machine learning algorithms.
What we learned
We learned a lot about financial analysis in building the stonks web app. We also learned a lot about deep learning techniques as we tried to implement that in our analysis.
What's next for Stonks
For stonks, we plan to further improve of UX as well as improving our testing and training data sets to give us a clearer picture of analysis.