Cognitive Identity Management: Synthetic Data, Risk and Trust

Synthetic, or artificial data is used in security applications such as protection of sensitive information, prediction of rare events, and training neural networks. Risk and trust are assessed specifically for a given kind of synthetic data and particular application. In this paper, we consider a more complicated scenario, – biometric-enabled cognitive cognitive biometric-enabled identity management, in which multiple kinds of synthetic data are used in addition to authentic data. For example, authentic biometric traits can be used to train the intelligent tools to identify humans, while synthetic, algorithmically generated data can be used to expand the training set or to model extreme situations. This paper is dedicated to understanding the potential impact of synthetic data on the cognitive checkpoint performance, and risk and trust prediction.