Sentiment Analysis Django Application

Overview

This Django application analyzes user-provided text to classify it as either positive or negative using a sentiment analysis model based on Long Short-Term Memory (LSTM) neural network. The model is trained on the IMDb movie reviews dataset.

Deployment

The application is deployed on the Render platform, and you can access it at https://analyzer-6fx2.onrender.com/.

Getting Started

To use the sentiment analysis functionality locally, follow the steps below:

1. Clone the repository:

   git clone https://github.com/ibrahim99035/ReviewAnalyzer.git
   cd ReviewAnalyzer

2. Install the required Python packages:

   pip install -r requirements.txt

3. Run the Django development server:

   python manage.py runserver

4. Access the application at http://localhost:8000/ in your web browser.

Model Training and Saving

The sentiment analysis model is trained using Keras and TensorFlow. The training code can be found in the train_model.py file. To train and save the model, run the following command:

python train_model.py

The trained model will be saved in the Model/sentiment_analysis_model.h5 file.

Model Loading and Prediction

To load the saved model and make predictions, the LSTM_Process class in lstm_process.py is used. This class includes a make_prediction method that takes a text statement as input and returns the predicted sentiment (positive or negative).

from lstm_process import LSTM_Process

lstm_processor = LSTM_Process()
prediction = lstm_processor.make_prediction("Your text statement goes here.")
print(prediction)

Django Views

The main functionality of the application is implemented in the views.py file. The home view uses the LSTM_Process class to make sentiment predictions based on user input.

Contributing

Feel free to contribute to the development of this sentiment analysis application. Fork the repository, create a new branch for your changes, and submit a pull request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

---