🎬 Sentiment Analysis using Simple RNN

Deep Learning based Movie Review Sentiment Classification using TensorFlow, Keras & Streamlit

⭐ Predict whether an IMDB movie review is Positive or Negative using a Recurrent Neural Network.

---

📌 Overview

This project implements a Natural Language Processing (NLP) pipeline using a Simple Recurrent Neural Network (Simple RNN) to classify IMDB movie reviews into:

• 😊 Positive Review
• 😞 Negative Review

The application is deployed with Streamlit, allowing users to enter their own movie reviews and instantly receive sentiment predictions along with confidence scores.

---

✨ Features

🤖 Deep Learning Model

• Simple RNN Architecture
• Word Embedding Layer
• Binary Sentiment Classification
• TensorFlow / Keras Implementation

---

🧠 NLP Pipeline

• Text Preprocessing
• Lowercase Conversion
• Tokenization
• Word Index Encoding
• Sequence Padding
• IMDB Dataset Vocabulary

---

🌐 Streamlit Application

• Interactive UI
• Real-time Prediction
• Confidence Score
• User-friendly Interface
• Fast Inference

---

🛠 Tech Stack

Technology	Purpose
Python	Programming Language
TensorFlow	Deep Learning
Keras	Neural Network API
Streamlit	Web Application
NumPy	Numerical Computing
IMDB Dataset	Movie Review Dataset

---

📂 Project Structure

Sentiment-Analysis-Simple-RNN-Project/
│
├── app.py                    # Streamlit Application
├── embedding.ipynb           # Word Embedding Experiments
├── simplernn.ipynb           # Model Training
├── prediction.ipynb          # Prediction Notebook
│
├── simple_rnn_imdb.h5        # Trained Model
│
├── requirements.txt
├── README.md

---

🧠 Model Architecture

                 Input Review
                       │
                       ▼
               Text Preprocessing
                       │
                       ▼
                 Word Encoding
                       │
                       ▼
               Sequence Padding
                       │
                       ▼
             Embedding Layer (128)
                       │
                       ▼
            Simple RNN Layer (128)
                       │
                       ▼
          Dense Layer (Sigmoid)
                       │
                       ▼
       Positive / Negative Prediction

---

⚙ Model Summary

Layer	Output Shape
Embedding	(None, 500, 128)
SimpleRNN	(None, 128)
Dense	(None, 1)

Total Parameters: 1,313,027

---

📊 Dataset

Dataset Used:

IMDB Movie Reviews Dataset

Dataset Characteristics

• 50,000 Movie Reviews
• Binary Classification
• Positive Reviews
• Negative Reviews
• Pre-tokenized Vocabulary
• Maximum Vocabulary Size: 10,000 Words

---

🔄 Workflow

Movie Review
      │
      ▼
Text Cleaning
      │
      ▼
Lowercase Conversion
      │
      ▼
Tokenization
      │
      ▼
Word Encoding
      │
      ▼
Padding Sequences
      │
      ▼
Embedding Layer
      │
      ▼
Simple RNN
      │
      ▼
Sigmoid Output
      │
      ▼
Positive / Negative

---

🚀 Installation

Clone Repository

git clone https://github.com/Kirisaki00/Sentiment-Analysis-Simple-RNN-Project.git

Move into Project

cd Sentiment-Analysis-Simple-RNN-Project

Create Virtual Environment

python -m venv venv

Activate Environment

Windows

venv\Scripts\activate

Linux / macOS

source venv/bin/activate

Install Dependencies

pip install -r requirements.txt

Run Streamlit

streamlit run app.py

or

python -m streamlit run app.py

---

💻 Usage

1. Launch the Streamlit application.

2. Enter a movie review.

Example:

This movie was absolutely amazing. The acting was incredible and the story kept me engaged throughout.

3. Click Classify

4. View

• Sentiment
• Prediction Score

---

📸 Screenshots

🏠 Home Screen

---

📈 Prediction Result

---

📊 Sample Predictions

Review	Prediction
This movie is amazing!	😊 Positive
Worst movie ever made.	😞 Negative
Fantastic acting and direction.	😊 Positive
Complete waste of time.	😞 Negative

---

📈 Future Improvements

• 🔥 LSTM Model
• 🚀 GRU Model
• 🤖 Transformer-based Sentiment Analysis
• 🌍 Multi-language Support
• 📊 Attention Mechanism
• ☁ Cloud Deployment
• 🐳 Docker Support
• 🔑 User Authentication

---

🤝 Contributing

Contributions are always welcome!

1. Fork the repository

2. Create a feature branch

git checkout -b feature-name

3. Commit changes

git commit -m "Added awesome feature"

4. Push changes

git push origin feature-name

5. Open a Pull Request

---

📜 License

This project is licensed under the MIT License.

---

👨‍💻 Author

Anupam (Kirisaki)

GitHub

https://github.com/Kirisaki00

---

⭐ If you found this project useful...

Give this repository a ⭐

Made with ❤️ using TensorFlow, Keras & Streamlit