GIT-CERCS-11-05
Apurva Mohan, Douglas M. Blough, Tahsin Kurc, Andrew Post, Joel Saltz,
Detection of Conflicts and Inconsistencies in Taxonomy-based Authorization Policies
The values of data elements stored in biomedical databases often draw from biomedical ontologies. Authorization rules can be defined on these ontologies to control access to sensitive and private data elements in such databases. Authorization rules may be specified by different authorities at different times for various purposes, and as such policy rules may conflict with each other, inadvertently allowing access to sensitive information. Detecting policy conflicts is non- trivial because it involves identification of applicable rules and detecting conflicts among them dynamically during execution of data access requests. It also requires dynamically verifying conformance with required policies and logging relevant information about decisions for audit. Another problem in biomedical data protection is inference attacks, in which a user who has legitimate access to some data elements is able to infer information related to other data elements. This type of inadvertent data disclosure should be prevented by ensuring policy consistency; that is, data elements which can lead to inference about other data elements should be protected by the same level of authorization policies as the other data elements. We propose two strategies; one for detecting policy consistencies to avoid potential inference attacks and the other for detecting policy conflicts. We have implemented these algorithms in Java language and evaluated their execution times experimentally.