Conceptual Structures for Inference Control

Due to the complexity of the infer ence problem we need several approaches to handle the problem. In Chapter 13 we discussed securityconstraint processing. In particular, we described the design of a MLS/DBMS that manages various types of security constraints. We stated the need to process security constraints during database design, database updates, and query operations. However, before the database is designed, we need to examine the application and see if there are potential violations to security. Essentially we need semantic data models such as semantic nets and conceptual graphs to model an application and reason about the application. This chapter attempts to examine the use of conceptual structures for inference control.