In this project, we collected 38,000 tweets, cleaned data, completed an EDA, analyzed sentiment using BERT, modeled topics using Latent Dirichlet Allocation (LDA), and effectively communicated our findings.
