Sandbox‎ > ‎Archive‎ > ‎IPT 2008-09‎ > ‎Tony's page‎ > ‎Tony's Updates‎ > ‎

Notes - Artificial Intelligence

posted Jun 15, 2009, 3:31 PM by Unknown user   [ updated Jun 15, 2009, 3:38 PM by Eddie Woo ]


How Expert Systems Work

Expert Systems have a KNOWLEDGE BASE of rules (domain-specific, heuristics) that gives situations and hence inform decisions.


1. Can handle more data more efficiently - instantaneously.
2. 'Sleep is for the weak'.
3. Availability - can serve multiple users simultaneously. Remains available past human lifetime.
4. Consistency - same decisions can be made without any inconsistencies.
5. There is a clear logic - comes out of preprogrammed rules.
6. They never forget things when making decisions.


1. They lack common sense and intuition.
2. Cannot make creative responses.
3. Limited by what it is given (the knowledge). It is, in essence, limited to its knowledge base. Since it cannot learn it needs to be constantly updated.
4. Does not adjust or adapt to certain situations.

Remember: Expert systems are designed to mimic a human expert. So why would you mimic a human expert when there already is a human expert?