Inspiration
Making money. Just kidding.
We thought using the Symbl API would be fun and using the sentiment analysis feature on stocks was the first good idea that came to mind.
What it does
Stock Stalk is a Java application that allows users to perform sentiment analysis on any stock using Symbl.ai, Yahoo Finance, and News API.
Simply search for a stock ticker and allow Stock Stalk to provide you with its price, individual sentiment score of recent news articles, and average sentiment score. Sentiment scores range from -1.0 (negative) to 1.0 (positive).
You can also save your favorite stocks for easy access on the homepage!
How we built it
Stock Stalk was built entirely using Java.
Three APIs were used:
- Yahoo Finance API to get stock information
- News API to get the most recent news articles about a stock to run sentiment analysis on
- Symbl API to run sentiment analysis on news articles
(Due to Symbl.ai only allowing for two concurrent jobs with a free account, only the two most recent news articles are used for the sentiment analysis. Upgrade your plan to get even more!)
The UI was created using Java Swing and designed using Figma.
Challenges we ran into
Getting the APIs to work with Java was the first challenge we ran into. We actually started our project with Java 1.8, but quickly realized it would be much easier with a newer version of Java.
Another challenge we ran into was creating the UI. Java Swing is not the easiest thing to learn and make good-looking UIs with so there were definitely many bugs and Google searches.
Lastly,
Accomplishments that we're proud of
As we had practically no past Java experience, finishing this project using only Java is an accomplishment we are extremely proud of. We are proud that we stuck with it and worked through the many challenges encountered along the way.
What we learned
A lot about Java, but even more about teamwork and planning.
What's next for Stock Stalk
We plan to add more sources to run sentiment analysis on, such as Twitter, Reddit, etc.
Some other features to add are: ticker autocomplete, using another UI framework, and adding even more stock information.
Log in or sign up for Devpost to join the conversation.