The Idea 💡
I’m sitting on about 300GB of file chaos scattered across my disks. I need a way to bring order to the mess. The goal is to cluster these files in a way that minimizes human effort in organizing them.

• Images and videos should be categorized using geotags and capture date metadata.
• Documents and other files should be categorized based on content, file type, and creation date.

The Plan 🎬
I plan to use a clustering algorithm to intelligently group similar files. A K-Means algorithm seems like a good starting point - assuming the metadata and tags can be encoded in a way that makes file-to-file distances meaningful.

The Code ⚗️
It doesn’t exist yet. When it does, expect it to be highly experimental. Please ensure you have a 3-2-1 backup strategy in place before running it.