https://github.com/rushilpatel21/redactify
Redactify is an efficient data redaction tool that secures sensitive text using advanced NLP and rule-based methods. It combines transformer-based NER, regex, and Presidio analysis to detect and mask personal information through full redaction or partial masking—ensuring compliance while preserving data utility.
https://github.com/rushilpatel21/redactify
anonymization data-redaction flask nlp pii presidio privacy pseudonymization python redactify regex transformers
Last synced: 3 months ago
JSON representation
Redactify is an efficient data redaction tool that secures sensitive text using advanced NLP and rule-based methods. It combines transformer-based NER, regex, and Presidio analysis to detect and mask personal information through full redaction or partial masking—ensuring compliance while preserving data utility.
- Host: GitHub
- URL: https://github.com/rushilpatel21/redactify
- Owner: rushilpatel21
- License: apache-2.0
- Created: 2025-02-21T13:40:27.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2025-03-01T13:23:13.000Z (3 months ago)
- Last Synced: 2025-03-01T14:26:52.937Z (3 months ago)
- Topics: anonymization, data-redaction, flask, nlp, pii, presidio, privacy, pseudonymization, python, redactify, regex, transformers
- Language: JavaScript
- Homepage: https://redactify.vercel.app
- Size: 392 KB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Redactify - Advanced PII Anonymization Platform
Redactify is a comprehensive solution for detecting and anonymizing personally identifiable information (PII) in text documents. The application combines a modern React frontend with a powerful Python Flask backend to provide an intuitive and effective PII redaction service.
### PS : Still figuring out a way to host server for free. (There are some memory constraints on service providers like render)
## 🔍 Overview
Redactify helps organizations comply with data privacy regulations by identifying and removing sensitive information from text. The platform uses a sophisticated multi-method detection approach combining machine learning models, rule-based patterns, and Microsoft's Presidio Analyzer to achieve high accuracy PII detection.
### Project Architecture
```
redactify/
├── client/ # React frontend application
│ ├── public/
│ ├── src/
│ ├── .env
│ └── package.json
│
├── server/ # Flask backend API
│ ├── server.py # Main API entry point
│ ├── .env
│ └── requirements.txt
│
└── README.md # This file
```
## ✨ Key Features
### Detection Capabilities
- **Multi-Method Detection**: Combines machine learning (Hugging Face NER), rule-based patterns (regex), and Presidio Analyzer
- **13+ PII Types**: Identifies personal names, organizations, locations, email addresses, phone numbers, credit cards, SSNs, IP addresses, URLs, dates, passwords, API keys, and roll numbers
- **Customizable Confidence Thresholds**: Configure detection sensitivity based on your needs
### Anonymization Options
- **Full Redaction**: Replace PII with categorized placeholders (e.g., `[PERSON-611732]`)
- **Partial Redaction**: Preserve some characters while masking others for better context retention
- **Selective Anonymization**: Enable/disable specific PII types for targeted redaction
### User Experience
- **Intuitive Interface**: Clean, responsive design works across devices
- **Real-time Feedback**: Toast notifications provide operation status
- **Animated UI**: Smooth transitions make interaction pleasant
- **Performance Optimizations**: Concurrent processing for faster results
## 🛠 Technologies Used
### Frontend (Client)
- React 18.2.0
- Framer Motion (animations)
- React Icons
- SweetAlert2 (notifications)
- CSS3 with custom styling
### Backend (Server)
- Python 3.9+
- Flask web framework
- Microsoft Presidio Analyzer
- Hugging Face Transformers (BERT NER model)
- Regular expressions for pattern matching
- Thread pool for concurrent processing
## 🚀 Installation & Setup
### Prerequisites
- Node.js (v16+)
- Python (v3.9+)
- npm or yarn
- pip
### Clone the Repository
```bash
git clone https://github.com/rushilpatel21/Redactify.git
cd Redactify
```
### Backend Setup
```bash
# Navigate to server directory
cd server
# Create and activate a virtual environment
python -m venv venv
source venv/bin/activate # On Windows, use `venv\Scripts\activate`
# Install dependencies
pip install -r requirements.txt
# Run the server
python server.py # Runs on http://localhost:8000 by default
```
### Frontend Setup
```bash
# Navigate to client directory
cd client
# Install dependencies
npm install
# Create .env file
echo "VITE_BACKEND_BASE_URL='http://localhost:8000'" > .env
# Start development server
npm run dev # Access at http://localhost:5173 by default
```
## 🔧 Environment Configuration
### Backend
The server uses environment variables for configuration:
- `PORT`: Server port (default: 8000)
- `CONFIDENCE_THRESHOLD`: Detection confidence threshold (default: 0.6)
### Frontend
The client uses a `.env` file for configuration:
- `VITE_BACKEND_BASE_URL`: Backend API URL (default: 'http://localhost:8000')
## 📝 API Documentation
### Anonymize Endpoint
```
POST /anonymize
```
#### Request Format
```json
{
"text": "The input text containing PII to anonymize",
"options": {
"PERSON": true,
"ORGANIZATION": true,
"LOCATION": true,
"EMAIL_ADDRESS": true,
"PHONE_NUMBER": true,
"CREDIT_CARD": true,
"SSN": true,
"IP_ADDRESS": true,
"URL": true,
"DATE_TIME": true,
"PASSWORD": true,
"API_KEY": true,
"ROLL_NUMBER": true
},
"full_redaction": true
}
```
#### Response Format
Success:
```json
{
"anonymized_text": "The redacted text with PII anonymized"
}
```
Error:
```json
{
"error": "Description of the error that occurred"
}
```
## 📋 Usage Examples
### Example 1: Full Redaction
**Input:**
```
This agreement is made between Generic & Associates (email: [email protected], phone: 555-123-4567) and Mr. John Smith (SSN: 123-45-6789).
```
**Output:**
```
This agreement is made between [ORGANIZATION-0458a5] (email: [EMAIL_ADDRESS-8eb1b5], phone: [PHONE_NUMBER-ca71de]) and Mr. [PERSON-611732] (SSN: [SSN-1e8748]).
```
### Example 2: Partial Redaction
**Input:**
```
Please contact John Smith at [email protected] or 555-123-4567.
```
**Output:**
```
Please contact Jo*****ith at jo****ith@*******.com or 55*******567.
```
---
© 2025 Redactify. All rights reserved.