Explanations and Trustworthiness in Salary

Project Overview

This repository contains the code and documentation for the project "Explanations and Trustworthiness in Salary Analysis Classification: A case study" conducted as part of the course "Explainable and Trustworthy Artificial Intelligence". The project explores the challenges of salary analysis classification, focusing on sensitive attributes such as gender, race, and age, and emphasizes the importance of explainable and trustworthy machine learning models.

Authors

• Valentina Ortega Pinto
• Martín Romero Romero

Repository Structure

• Data_Exploration_and_XAI.ipynb: Jupyter notebook containing data exploration and application of explainability methods.
• TAI.ipynb: Jupyter notebook focusing on exploring bias in the dataset and minimizing bias in trained models.
• data/: Directory containing the dataset used in the project.
• README.md: This file providing an overview of the project and instructions for running the code.

Running the Code

To run the code, follow these steps:

1. Clone the repository to your local machine.
2. Ensure you have Jupyter Notebook installed along with required dependencies such as Python libraries (e.g., pandas, numpy, scikit-learn).
3. Open and run the Jupyter notebooks Data_Exploration_and_XAI.ipynb and TAI.ipynb in the respective order.

Acknowledgments

• The dataset used in this project was obtained from Kaggle.