Chapter 7 - Rule-Based Systems

Abstract

This chapter provides an overview of the rule-based systems. Rule-based systems are also sometimes known as knowledge-based systems and expert systems. The major components of a rule-based system are the knowledge base that contains facts as well as rules, and the inference engine. The inference engine is the part where the reasoning takes place: where input information is combined with the rules and facts in the knowledge base to make decisions and construct new information. The rules usually come from humans via a user interface, most commonly a graphic user interface (GUI). The a priori facts which are built into the system also enter the system in this way. The process of getting knowledge from humans to put into machines is called knowledge elicitation. In expert systems, which attempt to capture and emulate some area of human expertise, knowledge elicitation is very important. This chapter concentrates on some of the particular issues relevant to rule-based systems irrespective of their application. This chapter also focuses on small rule-based systems which would be embedded in a system, rather than the larger rule-based systems. This chapter discusses basic concepts related to knowledge base, forward chaining, conflict resolution, and backward chaining. The chapter also explains diagnosis in rule-based systems.