3.11 A.I. and Expert Systems

what is the main function of expert systems?

help people make decisions

what are the advantages of expert systems?

-less expensive than human training
-can be copied
-always available
-contains accumulated knowledge

what two components do expert systems rely on?

-knowledge base: collection of facts about the system's domain
-inference engine: interprets/evaluates facts

what are the different modes of an expert system?

-knowledge acquisition: gains knowledge through people
-consultation: user asks a question, system answers
-explanation: system explains conclusion and reasoning

what is an expert system shell?

software used to create expert system

what is used to communicate with an expert system?

user interface

what are decision trees and how are they used?

-series of decisions/questions
-predetermined set of possible answers
-derives solutions by reducing the set of possible solutions

what is fuzzy logic based on (rather than T/F)?

degrees of truth

why is fuzzy logic more sensible to use within expert systems?

-natural language is not in absolute terms
-there are various states of truth (not just yes or no)
-fuzzy logic mimics brain functions

what are potential problems with expert systems?

-reliability: knowledge engineer didn't account for: ___
-the system is specific to one domain
-it is only applicable where info can be reduced to a set of computational rules

the expert system is a branch of:

artificial intelligence

what makes artificial intelligence human-like?

-uses past experiences to make more informed decisions
-builds on previous knowledge