Inspiration

One of our favorite YouTubers Michael Reeves is a genuinely crazy person, and he touched upon this idea of using social media posts to influence stock market investment decision. We know many people who are struggling with the stock market, so we decided to expand upon Michael Reeves' genius, and teach everybody how to... make money.

What it does

With a simple press of a button, users can see what stocks to invest in based on posts on a stock market-dedicated Reddit page.

How we built it

We constructed a model based on a dataset we found on Kaggle. Then we made a NodeJS website on Repl and downloaded the model onto there. We then made a separate Reddit scraping algorithm which returns a txt file for us to upload on the repl, to print the results on the website. The results consist of the Reddit headlines related to the stocks, and a determination on whether or not they would return a profit.

Challenges we ran into

We tried to use a Twitter Stock Market Sentiment Analysis model, but that failed with the model-creating code we had. So we switched to a simple Twitter Sentiment Analysis model. We also had a bit of trouble with the Reddit scraping, especially finding out which posts were appropriate to output the correct sentiment. We also had trouble connecting a Python backend to a NodeJS frontend repl, hence the use of a txt file. The predictions on the Reddit comments also took forever to run since there were so many of them, so I had to decrease the size of the headlines/comments that were being run through the prediction model so it could ouput faster.

Accomplishments that we're proud of

We are proud of getting the model to produce a decent accuracy and being able to make a Reddit scraper.

What we learned

We learned how to do sentiment analysis as none of us have used it before as well as how to scrape a website. We also learned how to make everyone rich by using machine learning.

What's next for Make Money

Make more money (Also train the model for a higher accuracy and find a way to incorporate Python into the repl)

Share this project:

Updates