Our main inspiration came from competing in stock market competitions. One of the best ways to find an edge was effectively manipulating a portfolio around up and coming news. In order to make this process more efficient myStocks serves as a tool to immediately compile reaction worthy news and sort between that which is positive and negative.

What it does

We scrape the top news sites around the globe to compile an in-depth list of articles for each stock in your portfolio. Through machine learning, we analyze the composition of each article and then refine the list to the top 6 most reaction-worthy positive and negative links. Users can follow the links to the articles provided for a comprehensive look into what the future holds for their investment portfolio!

How we built it

The web platform was built using a mixture of HTML and Javascript, supported by Google's Firebase API. The web scraper was built in Java and the system with which we used to rate articles - the "sentiment analyzer" - was built in Python.

Challenges we ran into

Bringing Firebase into our project as well as coordinating the different parts of the application proved to be difficult. In particular, integrating the web crawler and sentiment analyzer into the web interface was quite challenging. The storage and handling of data within the context of the web interface was something that was new to us.

Accomplishments that we're proud of

Our project involved a number of skills and technologies that many of us were new to or unfamiliar with. While the learning curve was steep, we are proud of what we were able to learn in such a short period of time. We are also proud of the efficiency as well as effectiveness of our web scraper and sentiment analyzer.

What we learned

Aside from a slew of new technologies, we also learned how to stay calm under stress and make things work no matter what. We also got the wonderful opportunity to get familiar with the hard floors : )

Share this project: