StockEye

Inspiration

We wanted to design a reliable method for predicting which stocks will be successful and to help investors make more educated decisions when buying/selling stocks.

What it does

This web application prompts the user to enter the ticker for a stock they would like to know more about. The website displays sentiment information about the stock from twitter posts, along with a graph showing the distribution of such sentiment information. There is also a recommendation if to buy the stock based off the given sentiment.

How we built it

We used html, css, and d3js in the frontend to give the user a clean experience. We used flask in the backend to provide the data.

Challenges we ran into

We had no previous experience in networking, so it was difficult to set up the flask server and send the predictions from the machine learning model back to the browser without running into API restrictions. This lead us to not being able to use a neural network, which was our previous plan, to predict whether or not to buy, sell or hold the stock.

Accomplishments that we're proud of

Successfully connecting the backend the the frontend.

What we learned

We learned how to use flask, keras, json, d3js, and how to connect these elements together to make a working application.

What's next for StockEye

We plan to analyze more sources for sentiment data, such as social media platforms. We also plan to incorporate a neural network to give users better stock predictions along with a wider set of training data, something we significantly lacked prior.