Sentiment Analysis System for Movie Reviews

I developed a complete Sentiment Analysis System that classifies IMDB movie reviews as positive or negative.

The system includes:

1-Preprocessing pipeline: case normalization, punctuation removal, stopword filtering, tokenization, and optional stemming/lemmatization.

2-Feature extraction using TF-IDF, saved for consistent inference.

3-Multiple trained models: Naive Bayes, Logistic Regression, Linear SVM, SVM (sigmoid kernel), and Random Forest.

4-Graphical User Interface (GUI) that allows users to input text or load a file, select a model, and get predictions instantly.

5-Batch testing scripts to validate models and run experiments.

How it was built:

The pipeline was implemented in Python using scikit-learn for machine learning and Tkinter for the GUI. Models and vectorizer were serialized with pickle for reuse. The dataset used is the IMDB polarity dataset, split into positive and negative reviews.