This project focuses on developing sentiment classification models for IMDB reviews, providing a beginner-friendly introduction to NLP models. Using shallow RNN, unidirectional LSTM, and bidirectional LSTM architectures, it incorporates GloVe embeddings and evaluates model performance with metrics like accuracy, precision, recall, and F1 score. The goal is to understand the impact of LSTM layers on model effectiveness, gain hands-on experience in NLP, and build a foundational skill set that is beneficial for working on larger language models (LLMs) and more advanced NLP projects in the future.
The notebook showcases the implementation on a code level: Click Here
This report provides details explaination of our work: Click Here
