IMDB Sentiment Analysis with CNN

Overview

This project performs sentiment analysis on IMDB movie reviews using a Convolutional Neural Network (CNN).

• Implemented in Python with TensorFlow/Keras.
• Redesigned from a previous Google Colab version to run locally without Colab.
• Includes data preprocessing, model training, evaluation, and sample predictions.
• Structured for easy understanding and reproducibility.

Files

File	Description
train.py	Loads data, trains the CNN, and evaluates accuracy
model.py	Defines the CNN architecture
preprocess.py	Loads IMDB dataset, tokenizes and pads sequences
requirements.txt	List of Python packages needed
.gitignore	Excludes unnecessary files (e.g., .idea/, CSV)
randum_30line.py	sample dataset

Dataset

• Uses IMDB Reviews dataset from TensorFlow Datasets.
• The raw CSV dataset is NOT included due to licensing.

Sample Dataset

A small random subset (30 reviews) of the IMDB Dataset is included for testing purposes. Full dataset is downloaded automatically by train.py.