1. introduction to knowledge-based intelligent systems(summary / questions for review / references)
2. rule-based expert systems
3. uncertainty management in rule-based expert systems
4. fuzzy expert systems
5. frame-based expert systems
6. artificial neural networks
7. evolutionary computation
8. hybrid intelligent systems
9. knowledge engineering and data mining
2. rule-based expert systems
production rules are represented as IF (antecedent) THEN (consequent) statements. A production rule is the most popular type of knowledge representation. Rules can express relations, recommendations, directives, strategies and heuristics.
the expert system development team:
a rule-based expert system has five basic components:
the complete structure of a rule-based expert system:
In developing rule-based expert systems, shells are becoming particularly common. An expert system shell is skeleton expert system with the knowledge removed. To build a new expert system application, all the user has to do is to add the knowledge in the form of rules and provide relevant data. Expert system shells offer a dramatic reduction in the development time of expert systems.
Expert systems can deal with incomplete and uncertain data and permit inexact reasoning. However, expert systems can make mistakes when information is incomplete or fuzzy.
There are two principal methods to direct search and reasoning: forward chaining(data-driven reasoning) and backward chaining(goal-driven reasoning) inference techniques. If an expert first needs to gather some information and then tries to infer from it whatever can be inferred, choose the forward chaining inference engine. If your expert begins with a hypothetical solution and then attempts to find facts to prove it, choose the backward chaining inference engine. Or maybe you can use both of them for your special problem.
If more than one rule can be fired in a given cycle, the inference engine must decide which rule to fire. A method for deciding is called conflict resolution. We can use those conflict resolution methods: rule order, rule priority, rule add time, rule specific(processe more information), and We can use Metaknowledge(the knowledge about knowledge), in rule-based expert systems, metaknowledge is represented by metarules.
Rule-based expert systems have the advantages of natural knowledge representation, uniform structure, separation of knowledge from its processing, and coping with incomplete and uncertain knowledge.
Rule-based expert systems also have disadvantages, especially opaque(不透明的) relations between rules, ineffective search strategy, and inability to learn.
