social-media-PII-scrubber

Python utility for Parsing and scrubbing PII from social media dumps Credit: Derived from jwindha1/sm-parser (which was forked from cspang1 & ndo3)

Extending functionality to include:

• Facebook (v2 Schema)
• Instagram (v2,v3 Schemas)
• Snapchat
• TikTok

Codebase improvements:

• Classes to consolidate multi-use functions and data management
• Lower temp memory demands: Zip files are not unzipped/destroyed
• GUI: Simple input form for input values & launch parser
• Progress bars for longer running processes & Options to skip media (face blurring) sequence

Installation sequence (for Windows)

• Download Visual Studio
• Install cmake
• Install dlib (for face_recognition)
• python -m textblob.download_corpora (this will download ntlk_data, probably to your home/AppData/roaming folder)
• Relocate nltk_data into your virtual environment .venv folder (path should be ./.venv/nltk_data)
• Note: The Pyinstaller process (from launch_smparser.spec) requires numerous 'hook' files to collect the hidden imports and data files. These are included in the ./hooks folder.

Release Notes

• v0.5 - IG Parameter paths read from data_mapping.json (editable outside executable); Face blur settings
• v0.4 - Changed name detector/library to textblob
• v0.3 - TTParser modified to suit data anonymity
• v0.2 - SCParser added; New feature: Alias a custom list of values; Fix: Update Starting Date as duration is changed