Approach to Inference Disclosure Detection in Secure Databases
Xiao Jun-mo · Journal of Changchun Post and Telecommunication Institute · 2005
The problem of inference disclosure in secure databases that occurs when a generic user circles security mechanism and obtains sensitive data in the database combining the acquired data and application knowledge of the database. To resolve the problem, an approach to inference disclosure detection at data level was proposed. The approach was expressed in form of formal system and detected inference disclosure with logic-based techniques based on database constraints and data user acquired. It is proved theoretically and showed by testing result that the approach can obtain material implications from the acquired data completely and effectively, and its correctness rate of detection can get up to 81.3% with those material implications.