We provide support for the removal of bias in datasets or models generated from data.
AI techniques based on big data and algorithmic processing are increasingly used to guide decisions in important societal spheres, including hiring decisions, university admissions, loan granting, and crime prediction. The discriminative impact of AI-based decision-making to certain population groups has been already observed in a variety of cases and the need to move beyond traditional AI algorithms optimized for predictive performance has been identified. We offer support and advice on how standards of unbiased attitudes and non-discriminatory practices can be incorporated in big data analysis and algorithm-based decision-making.
SPECIAL ACCESS CONDITIONS
Possible use cases are in all areas, where personal data is used to train a decision making model. These are particularly vulnarable to bias resulting from the training data. One application example are financial services, where credit limit models can be biased (directly or indirectly) by home address, age, gender, etc.