We saw the effect of twitter blue subscription tweets on the stocks of corporations such as Lockheed Martin among others which gave us the idea.

What it does

It looks at the tweet and categorizes the sentiment and uses it to predict the effect of that on the stock prices of the corporation being impersonated

How we built it

We used numpy and Panda to create the ML model and used Django to tie it to a website.

Challenges we ran into

We ran into the challenge of not being familiar with making a NLP model.

Accomplishments that we're proud of

We were able to process 1,600,000 tweets to use them for training the model

What we learned

We learned that making a full-stack project can be an intensive and time-consuming process

What's next for WatchBot

Fleshing out the web-UI and making it work more effectively.

Share this project: