| # Bias Detection in AI-Assisted Hiring Systems | |
| A comprehensive dashboard for detecting and mitigating algorithmic bias in AI-driven recruitment. | |
| ## Features | |
| - **Disparate Impact Analysis**: Tracks selection rates across Gender and Race. | |
| - **Intersectional Equity Audit**: Analyzes bias at the junction of multiple demographic identities. | |
| - **Bias Simulation**: Predicts hiring outcomes for custom candidate profiles. | |
| - **Fairness Mitigation**: Integrated "Fair-Mask™" mitigation strategy using dynamic thresholding. | |
| ## Stack | |
| - **Backend**: Flask, Scikit-learn, Pandas | |
| - **Frontend**: Vanilla CSS (Glassmorphism), Chart.js | |
| - **Model**: Logistic Regression for hiring outcome simulation. | |
| ## How to Run | |
| 1. Install dependencies: | |
| ```bash | |
| pip install -r requirements.txt | |
| ``` | |
| 2. Train the model: | |
| ```bash | |
| python train_model.py | |
| ``` | |
| 3. Run the application: | |
| ```bash | |
| python app.py | |
| ``` | |
| 4. Access the dashboard at `http://127.0.0.1:5001` | |