Inspiration
We get inspiration by analyzing the dataset and looking at the information in the dataset that we could have. Then we decided to extract data about tweets text and likes numbers to help us for further analysis We hope to raise awareness for climate-related issues and also hope to maintain a healthy internet environment as well as retain the users in the platform for interactions.
How we built it
We use Bert Transformers for our NLP analysis
Accomplishments that we're proud of
We fit the model and get an accuracy rate of around 82%. That means we could accurately predict whether the tweets will be liked or not.
What's next for Climated-related twitter feedback analysis using Bert
We will present our output and try to dig deeper into our algorithm and model's application and contribution to climate system.
Built With
- https://github.com/psyduck572/tweets-bert-.git
Log in or sign up for Devpost to join the conversation.