Inspiration

We wanted to initially do something about making a semantics analyzer for financial news to analyze whether the news headlines is bullish or bearish.

What it does

Our project takes a twitter post and returns its assumed semantics.

How we built it

We built it using a custom made LSTM neural network and trained it on a database provided by huggingface

Challenges we ran into

We ran into a big problem with testing time as just training the model would take up to 20 minutes.

Accomplishments that we're proud of

We're proud of getting the nn up and running, despite it being extremely overfitted.

What we learned

We learned a lot about building custom networks.

What's next for Twitter Post Analytics

Probably get it up and running, and hopefully create an api access for the model so that it can be used to analyze twitter posts as a tool.

Built With

Share this project:

Updates