2011/12/03 by Faisal Kamiran, Toon Calders · 1 citation
Computer Science · #Artificial intelligence #Binary classification #Classifier (UML) #Computer science #Data mining #Data pre-processing #Imbalanced Data Classification Techniques #Labeled data #Machine Learning and Algorithms #Machine Learning and Data Classification #Machine learning #Noisy data #Pattern recognition (psychology) #Preprocessor #Resampling #Support vector machine #Test data
paper · pdf · doi:10.1007/s10115-011-0463-8
crossref issued 2011/12/03 · crossref published 2011/12/03 · crossref published-online 2011/12/03 · openalex publication_date 2011/12/03 · crossref created 2011/12/03 · crossref published-print 2012/10/01 · crossref deposited 2019/06/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06 · crossref indexed 2026/08/07
Recently, the following Discrimination-Aware Classification Problem was introduced: Suppose we are given training data that exhibit unlawful discrimination; e.g., toward sensitive attributes such as gender or ethnicity. The task is to learn a classifier that optimizes accuracy, but does not have this discrimination in its predictions on test data. This problem is relevant in many settings, such as when the data are generated by a biased decision process or when the sensitive attribute serves as a proxy for unobserved features. In this paper, we concentrate on the case with only one binary sensitive attribute and a two-class classification problem. We first study the theoretically optimal trade-off between accuracy and non-discrimination for pure classifiers. Then, we look at algorithmic solutions that preprocess the data to remove discrimination before a classifier is learned. We survey and extend our existing data preprocessing techniques, being suppression of the sensitive attribute, massaging the dataset by changing class labels, and reweighing or resampling the data to remove discrimination without relabeling instances. These preprocessing techniques have been implemented in a modified version of Weka and we present the results of experiments on real-life data.