Inspiration

Wanted to analyze how Twitter sentiments correlate to populace disposition on current discussions of interest.

What it does

Utilizes past tweet sentiments to determine the polarity of an input tweet.

How I built it

We used Python to interpret and analyze the data.

Challenges I ran into

Filler words negatively affected the success rate of our algorithm.

Accomplishments that I'm proud of

We have a 75-80% success rate on determining tweet polarity on test data.

What I learned

How to analyze polarity and sentiments of text.

What's next for Rice Datathon 2019

To develop a way to analyze historical context and cluster tweet polarity to determine how people feel on certain topics.

Built With

Share this project:

Updates