This project involves training a transformer model BERT on a dataset of 25,000 hotel reviews from Kaggle (Trip Advisor Hotel Reviews). The model predicts the sentiment of each review as positive, negative, or neutral.
The project was later extended to include traditional machine learning models such as Naive Bayes, Decision Tree, and Random Forest. These models were trained on the same dataset, and their performance was compared to the transformer-based model to analyze differences in accuracy and effectiveness.