An automated Privacy Engineering tool designed for the DACH region to detect and anonymize Personal Identifiable Information (PII) in German texts.

In the era of strict data privacy regulations (GDPR / DSGVO), automated data minimization is essential.

GDPR Guardian is a Python-based utility that leverages Natural Language Processing (NLP) and region-specific regex patterns to sanitize documents before they leave secure environments.
It is specifically engineered to handle German language nuances and Austrian/German formats.

• 🇦🇹 Region-Specific Detection: Accurate identification of Austrian IBANs (AT...) and Phone Numbers (+43).
• 🧠 AI-Powered Named Entity Recognition: Uses Spacy's de_core_news_sm model to detect German Names and Locations contextually.
• 📄 Multi-Format Support: Processes both plain text (.txt) and PDF documents (.pdf).
• 🔒 Privacy by Design: Implements pseudonymization placeholders (e.g., [PERSON_GDPR]) to maintain document structure while removing sensitive data.

1. Clone the repository:

git clone https://github.com/osmankaankars/GDPR-Guardian.git
cd GDPR-Guardian

2. Install dependencies:

pip install -r requirements.txt

3. Download the German Language Model:

python -m spacy download de_core_news_sm

Run the tool via command line by passing the target file:

python anonymizer.py kunde_wien.txt

Client: Hans Müller, Location: Wien, IBAN: AT89 3704 0044 0532 0130

Client: [PERSON_GDPR], Location: [LOCATION_GDPR], IBAN: [IBAN_REDACTED]

Osman Kaan Kars
Cybersecurity Engineer | Privacy Engineering Enthusiast

Connect with me on LinkedIn for specialized DACH region security projects.