We found that trading in the stock market is a laborious activity that involves tons of research. We want to find a way for a stock trader to efficiently and effectively get the information they need to make smart decisions in the stock market.

What it does

Our program reads an article involving a company and performs a sentiment analysis. The program then predicts the change in stock value for the company based on information from the article.

How we built it

All the code files for our program are organized in GitHub. Our group split up the work into three main components: UI, data collection, and AI. We found articles on the web concerning well known companies and traced their stock history during the week the article was published. The AI takes the data gathered to predict the stock price of a company for the next 7 days based on an article. The user interface is simplistic and organized for the user to conveniently use our service.

Challenges we ran into

Due of time constraints, we are unable to have our program be able to search on the web for articles that may influence stock. We were planning on using code online that performs a sentiment analysis in our project, but ran into difficulties when trying to implement it. Ultimately, we ended up coding our own simplistic sentiment analysis.

Accomplishments that we're proud of

We are proud to be able to make a working program within 6 hours. For most of us, this is our first Hackathon, and we have done more than we expected today.

What we learned

We learned how to effectively organize and manage our project using GitHub. We also gained insight on how the stock market works and how the media influences the success of a business. Furthermore, we learned about the importance of creating trust with our users.

What's next?

For our AI to become more accurate, much more training data must be fed to our program. We will also improve the algorithm behind our sentiment analysis. Moreover, we are hoping to add a working UI to pair with our backend.

Built With

Share this project: