Knowledge base systems are an integration of conventional database systems with artificial intelligence techniques. They provide inference capabilities to the database system by encapsulating the knowledge of the application domain within the database. Knowledge is the most valuable of all corporate resources that must be captured, stored, re-used and continuously improved, in much the same way as database systems were important in the previous decade. Flexible, extensible, and yet efficient knowledge base systems are needed to capture the increasing demand for knowledge-based applications which will become a significant market in the next decade. Knowledge can be expressed in many static and dynamic forms, the most prominent being domain objects, their relationships, and their rules of evolution and transformation. It is important to express and seamlessly use all types of knowledge in a single knowledge base system. This work presents in detail features that a knowledge base system should have in order to fulfill the above requirements.
It covers in detail the following topics: integration of deductive, production, and active rules in sequential database systems; integration and inter-operation of multiple rule types into the same knowledge base system; and parallel rule matching and execution, for deductive, production, and active rules, in parallel export, knowledge base, and database systems.