Discrimination-aware association rule mining for unbiased data analytics

Abstract

A discriminatory dataset refers to a dataset with undesirable correlation between sensitive attributes and the class label, which often leads to biased decision making in data analytics processes. This paper investigates how to build discrimination-aware models even when the available training set is intrinsically discriminating based on some sensitive attributes, such as race, gender or personal status. We propose a new classification method called Discrimination-Aware Association Rule classifier (DAAR), which integrates a new discriminationaware measure and an association rule mining algorithm. We evaluate the performance of DAAR on three real datasets from different domains and compare it..