1. Pre-process and clean the data. Remove HTTP links and usernames.
2. Train two models for classification
a. Fine-tune a pre-trained BERT model for sentiment analysis. You will use the BERT model provided by the HuggingFace library and add additional layers for classification.
b. Train a Naive Bayes Classifier using CountVectorizer
You can not use a model that is already trained on Twitter data or has inbuilt classification layers. The goal of the assignment is to familiarize yourself with the process of extending pre-trained models for your downstream task.
