Mitigating Gender Bias in Credit Scoring using Fair Machine Learning Techniques

Choudhary, Nilesh Kesharam (2025) Mitigating Gender Bias in Credit Scoring using Fair Machine Learning Techniques. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (659kB) | Preview
Preview	PDF (Configuration Manual) Download (775kB) | Preview

Abstract

Machine learning-based credit scoring systems are increasingly used to make high-stakes financial decisions, raising concerns that these systems can perpetuate historical social inequalities. A particularly sensitive issue is that of gender bias, where applications by people of similar financial standing can be treated differently because of embedded patterns in past data. This research investigates the issue of gender disparity in credit scoring predictions and examines the performance of a multi-stage fairness-aware machine learning pipeline to reduce this bias. The framework unites methods common to preprocessing (Reweighing, Feature Editing), in-processing (Exponentiated Gradient), and post-processing (Threshold Optimiser, Reject Option Classification) and evaluates them in terms of their effect on the performance and fairness evaluations. Using the methods of XGBoost and Logistic Regression and focusing on the evaluation of Statistical Parity Difference (SPD), Disparate Impact (DI), Equal Opportunity Difference (EOD), and traditional performance metrics like Accuracy and AUC. Results show that fairness interventions can be powerful for reducing gender disparity - improving SPD from -0.07 to 0.003 while keeping the competitive predictive ability high. The work contributes as a reproducible, end-to-end fairness pipeline and insights, hoping to contribute to the analysis of trade-offs between accuracy and fairness in fairness for financial decision-making systems.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Chikkankod, Arjun UNSPECIFIED
Uncontrolled Keywords:	Fairness; Machine Learning; Credit Scoring; Gender Bias; Fairlearn; AIF360; XGBoost; Logistic Regression; Responsible AI
Subjects:	H Social Sciences > HG Finance Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence H Social Sciences > HG Finance > Credit. Debt. Loans. Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	07 Sep 2026 10:08
Last Modified:	07 Sep 2026 10:36
URI:	https://norma.ncirl.ie/id/eprint/9852

Actions (login required)

View Item