A Nonmonotonic Typed Multilevel Logic for Multilevel Secure Data / Knowledge Base Management Systems - I

The study of databases through formal logic has not only enabled efficient proof procedures to be developed for query evaluation and integrity checking, but it has also resulted in the development of more intelligent and powerful database management systems. Despite all these advances, a logic for reasoning in a multilevel environment or a logic programming system for multilevel environments does not exist at present. As a result, multilevel secure database management systems lack several features that have been successfully incorporated into the conventional database management systems. These include constraint processing, deductive reasoning, and handling efficient proof procedures. Classical first order logic, being monotonic, is an inappropriate tool for formalizing concepts in multilevel databases. This is because, in a multilevel environment, it is possible for users at different security levels to have different views of the same entity. In other words, statements that are assumed to be true at one security level can very well be false at a different security level. Therefore, a special logic is needed for formalizing concepts in multilevel databases as well as to develop multilevel database systems. In our earlier paper we described a logic called Nonmonotonic Typed Multilevel Logic (NTML) for multilevel database applications. We also described various approaches to viewing multilevel databases through NTML. In this paper we continue with our discussion of the applications of NTML. In particular, the use of NTML as a programming language, issues on handliig negative information in multilevel databases, and approaches for integrity checking in multilevel database systems are described. Our work on NTML will be of significance to multilevel dataknowledge base applications in the same way logic programming has been to the development of damowledge base applications.