untitled

54 3 Rules and Inference

kinds of automated reasoners. Including rule-based systems, the main kinds

of automated reasoners are

1. forward-chaining rule engines,


2. backward-chaining rule engines,


3. theorem provers,


4. constraint solvers,


5. description logic reasoners,


6. business rule systems,


7. translators,


8. miscellaneous systems.

These are not mutually exclusive categories, and some systems support more

than one style of reasoning. We now discuss each of these categories in detail,

and then give a list of some of the available software for automated reason-

ing.

Summary

• Rule-based programming is a distinct style from the more common pro-
  cedural programming style.


• A rule consists of an antecedent and a consequent. When the antecedent
  is satisfied, the consequent is invoked (fired).


• Rule engines logically infer facts from other facts, and so are a form of
  automated reasoning system.


• There are many other kinds of reasoning system such as theorem provers,
  constraint solvers, and business rule systems.

3.2 Forward- and Backward-Chaining Rule Engines

Forward-chaining rule engines are the easiest rule engines to understand.

One simply specifies a set of rules and some initial facts. The engine then

fires rules as long as there are rules whose match condition is true. Of course,