Data Governance on EA Information Assets: Logical Reasoning for Derived Data

Today’s companies face increased pressure regarding compliance to legal obligations. Regulations for the financial sector such as Basel II and III, Solvency II, or the Sarbanes-Oxley-Act explicitly demand various requirements. Many of those requirements address the governance and management of information assets, such as data. Companies need to report and track their information architecture, and furthermore have to provide accountability and responsibility information on their data to, e.g., supervisory authorities. Additionally, the tracking of processed data becomes increasingly difficult since the software systems and their interactions throughout the enterprise are highly complex. This paper argues for a consistent and comprehensive assignment mechanism on data governance roles. Based on logical inferences, we are able to show how accountability and responsibility can be assigned throughout processed data. Thereby, we analyze the limitations of traditional logic, such as propositional logic, and exemplarily show how non-monotonic defeasible logic can be used to keep the assignment of roles on information assets consistent.