Have you ever thought of a better way to do role modelling?
The goal of this presentation is to show how Big Data analysis can be used to obtain a different and effective way of role modelling. Just like in traditional role mining, we take entitlement data from the source applications, but apply tools from the Big Data “world” and use these for analyzing the data.
By comparing accounts and entitlements by definition of their similarity, it is possible to make a visual presentation indicating logical groupings of accounts and entitlements, and use the visualizations to present the actual data to the business.
Hereby we can make a model showing what could be a proper role mining scheme for the actual entitlement data.
Key takeaways:
(1) a strict way to define account and entitlement similarity
(2) a proposal for visualization of logical groups of accounts and entitlements based on real data
(3) an analytical approach for combining a top-down and bottom-up approach in role mining
