Text Sentiment Analyzer (Naive Bayes)

This project is an end-to-end Machine Learning solution for classifying movie review text into two categories: Positive or Negative. It uses a trained Multinomial Naive Bayes model and is packaged with a runnable Gradio web interface for instant demonstration.

This project is a perfect example of a fast, client-ready ML deployment artifact.

---

🎯 Project Goal & Results

The primary goal was to build a quick and accurate sentiment classifier using classical ML techniques.

• Model Used: Multinomial Naive Bayes (MultinomialNB)
• Vectorizer: CountVectorizer (Bag-of-Words) with a maximum of 5,000 features.
• Dataset: IMDB Dataset of 50,000 Movie Reviews.
• Final Accuracy: 84.71%

⚙️ How to Run the Demo Locally

This project is designed for immediate execution, demonstrating a professional ML delivery pipeline.

Prerequisites

1. Python (3.7+)
2. Install required libraries using the provided requirements.txt file:

   pip install -r requirements.txt

Execution Steps

1. Download all files from this repository (the .ipynb, .py, .txt, and the two .pkl files).
2. Open your terminal or command prompt in the project directory.
3. Run the application script:

   python app.py

A link will appear in your terminal (usually http://127.0.0.1:7860). Open that link in your browser to interact with the model instantly!

---

📁 Repository Contents

File	Description	Purpose
Text_Sentiment_Analyzer.ipynb	The full Jupyter Notebook.	Documents the complete training, preprocessing, and evaluation process.
app.py	The main execution script.	Loads the model artifacts and launches the Gradio UI. The client-facing runnable file.
multinomial_nb_model.pkl	The saved, trained Naive Bayes model.	Enables instant predictions without time-consuming re-training.
count_vectorizer.pkl	The saved, fitted vocabulary (Vectorizer).	CRITICAL for converting new text input into the numerical format the model expects.
requirements.txt	List of all necessary Python dependencies.	Ensures easy, reproducible setup for any user.
README.md	This project overview.	Professional documentation and instructions.

🔗 Dataset Details

The dataset used for training this model (50,000 IMDB movie reviews) is publicly available:

• Source Link: IMDB Dataset of 50k Movie Reviews on Kaggle

---