Development of tools to detect dataset bias
Description Automated decision-making (ADM) systems may affect multiple aspects of our lives. In par- ticular they can result in systematic discrimination of specific population groups, in violation of the EU Charter of Fundamental Rights. One of the potential causes of discriminative be- havior, i.e. unfairness, lies in the quality of the data used to train such ADM systems.
Using a data quality measurement approach combined with risk management, both defined in ISO standards, the work focuses on balance characteristic to understand how balance of protected attributes in training data can be used to assess the risk of unfairness and discimination in ADM systems.
Required skills Good knowledge of data analysis techniques and tools , R or python
Deadline 30/06/2022 PROPONI LA TUA CANDIDATURA