RedactNLP: Redact Your PDF!

RedactNLP is a tool that allows you to automatically redact sensitive information from PDF documents using natural language processing and computer vision techniques.

Visit RedactNLP Spaces for the hosted demo on Huggingface Spaces.

How Redaction Works

1. PDF to Images: The PDF pages are converted into images.
2. Text Extraction: Using EasyOCR, text is extracted from the images.
3. Entity Identification: The "dslim/distilbert-NER" model classifies tokens in the extracted text to identify sensitive elements such as names, locations, or organizations.
4. Redaction: A non-recoverable mask is applied to all identified sensitive elements, ensuring that they cannot be recovered from the redacted document.

Features

• Automatic redaction of sensitive information from PDFs.
• Uses OCR to handle scanned or image-based PDFs.
• Leverages state-of-the-art NLP models for entity recognition.
• Ensures irreversible redaction of confidential information.

Installation

Clone the repository and install the required dependencies:

git clone https://github.com/mitanshu7/RedactNLP.git
cd RedactNLP
pip install -r requirements.txt

Usage

To redact a PDF:

python app.py

Then navigate to http://localhost:7860

Contributing

Feel free to open issues or submit pull requests if you would like to contribute to this project.

License

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