Inspiration
We wanted to find a new metric to help with investing. A family member of one of the team members writes articles for a financial institution, so we felt that it would be a good idea to evaluate such articles.
What it does
It first collects news headlines about a certain company and then runs sentiment analysis on those articles. Then it gets stock data for that same company. Both the sentiment and stock data is outputted in an internal API, where it is then displayed in a graph on the frontend.
How we built it
We use Node.js on the backend to get data from various public APIs. We then analyze that data and output the results as JSON. On the frontend the search page makes an AJAX request to the internal API and displays the results in a chart.
Challenges we ran into
Differing amounts of data made it hard to match up the sentiment and price data. Eventually we were able to create an algorithm that provided us with similar amounts of sentiment and price data (based on time).
Accomplishments that we're proud of
Creating a product that actually is relevant for investing. Many stocks actually show a strong correlation, which will help investors when deciding whether to invest.
What we learned
We learned a lot about asynchronous requests and JavaScript promises and callbacks. We also learned a lot about stock data and sentiment analysis.
What's next for Sentimentepede
We don't know yet!
Log in or sign up for Devpost to join the conversation.