Sentiment Analysis: Traditional NLP vs. Transformer Models

This project is designed to build and compare two distinct sentiment analysis systems using a dataset of user-written movie reviews. The primary goal is to classify each review into one of two categories: Positive or Negative (e.g., "This movie was absolutely amazing." -> Positive).

By implementing both a Traditional NLP Pipeline and a Transformer-Based Pipeline (using BERT), this project provides a comprehensive comparison of classical machine learning techniques against state-of-the-art deep learning approaches in terms of accuracy, computational speed, and the ability to understand nuanced language and context (such as sarcasm).