|
This tutorial provides only an overview of the essentials of knowledge
engineering as it applies to small rule-based systems.
Most of the books listed in the
references
provide a more comprehensive treatment. Here are some examples of additional
knowledge engineering topics you could learn about in these sources.
|
 |
Interviewing techniques. The most common method for capturing
expertise is the personal interview, and there are many techniques available
for planning, controlling and documenting the interview process to make it
as effective as possible.
|
|
Observation methods for acquiring expertise. In some cases it will be
efficient to observe an expert accomplishing the task that will be captured
in a knowledge base rather than asking the expert to describe it.
This might be done live or with videotape.
|
|
Knowledge analysis techniques. Graphical knowledge representation
devices such as flow charts, decision trees and decision tables are
often useful in capturing or refining expertise before it is converted into
the format required by the knowledge base. These knowledge representations
should be verified by the expert to assure that the knowledge
has been accurately captured.
|
|
Automated rule generation. When case data can be put into a
machine readable form, it is sometimes possible to use rule induction
software to mechanically generate if/then rules that are consistent with the data.
|
|
This concludes the Knowledge Engineering tutorial. If you
would like a more general overview of expert systems, the
Expert System Introduction is recommended. To see how rule-based
expert systems reason and deal with uncertain data take a look at the
Inference Methods and Uncertainty tutorial. If you have never run the
expert systems provided on this Web site,
Using eXpertise2Go's
Knowledge Bases is a good starting point.
|