Artificial Intelligence – Expert Systems

Artificial Intelligence – Expert Systems

In artificial intelligence, an expert system is a computer system that emulates the decision-making ability of a human expert.


A system like this is able to solve problems in a limited area but with the performance similar to that a man has.
It is also an example of knowledge-base system which is composed by:
– the knowledge-base,
– the inference engine.
In fact, they reach the goal using the knowledge base, better if it is represented with quantity and quality of data, and extracting always more information from these data until they reach the aim.

They are characterized by some properties:
– generality
– explicit representation of the knowledge
– reasoning mechanisms
– explanation mechanisms
– ability to work even with not-structured domains

Each expert system has to be able to:
– know very well the domain of application
– know how to use the knowledge to solve problems

Knowledge-base +
Inference Engine =
Knowledge-base system

We already said that an expert system is an example of a knowledge-base system. Let us have a deeper look at how these systems are built.

The knowledge-base is a simple set of statements that map the real world things, representing barely facts. Originally, they were flat assertions, but with the introduction of object-oriented programming languages, the real world was described using classes and instances: the assertions was replaced by the values of object instances.

The inference engine is an automated reasoning system that evaluate the current state, applies rules, and add new knowledge to the knowledge-base.


Mycin was an expert system in diagnosis, used to infer system malfunctions from observations.


The advantages are:

  • easy to maintain

The disadvantages are:

  • the knowledge acquistion problem
  • performance


Leave a Reply

Your email address will not be published. Required fields are marked *