Residential Mortgage Application Credit Scoring dataset

Residential Mortgage Application Credit Scoring dataset是由都柏林大学学院计算机科学学院创建的合成数据集，包含37,607条样本，用于研究机器学习中的偏见问题。该数据集模拟了住宅抵押贷款申请的信用评分，具有18个特征，包括年龄、收入等，其中年龄被选为敏感特征。数据集的创建过程涉及复杂的特征交互，旨在通过调整年龄和其他特征的阈值来模拟不同程度的偏见。该数据集主要用于评估和解决机器学习模型中的偏见和不公平问题，特别是在信用评分和贷款批准领域。

The Residential Mortgage Application Credit Scoring dataset is a synthetic dataset created by the School of Computer Science, University College Dublin. It contains 37,607 samples and is intended for research on bias in machine learning. This dataset simulates credit scoring for residential mortgage applications, with 18 features including age and income, among which age is selected as the sensitive feature. The dataset was developed through complex feature interactions, aiming to simulate varying degrees of bias by adjusting the thresholds of age and other features. It is primarily used to evaluate and mitigate bias and unfairness issues in machine learning models, particularly in the domains of credit scoring and loan approval.