Count-Based FairDeDup (CB-FDD)

Course project for EECS6320 Fairness and Bias in AI.
Contributors: Aiza Bajwa, Katherine Ling, Elham Razi.

This project is based on FairDeDup method proposed in this paper

Please find the main work in the notebooks folder.

How to run

1. Download the .ipynb files in the notebooks folder.
2. Open the .ipynb files in Google Colab.

Notes

1. Data Access: We assume you can access the MIMIC CXR embeddings and other required files (e.g., metadata, demographics).
2. Starting Point: We begin with the processed and merged dataset you get after running the pre-processing notebook (available on eClass).
3. File Paths: The notebooks reference file paths from our Google Drive setup. You may need to update the paths in the code to match wherever you upload your files.