|
This page is the root in a web of
pages that answers the question What is Pro/3? - in a brief top-down
way. Emphasis is made on explaining concepts,
rather than on explaining how to do things. The briefest possible description of the Pro/3 system is given in the
figure to the right:
The most basic concept in Pro/3 is the concept of a fact. Understanding facts is a prerequisite to understanding other Pro/3 concepts. |
|
|
|
|
|
What can Pro/3 be used for? Pro/3 can be used to build expert-systems, i.e. applications which can perform expert-like reasoning in a problem domain, for the purpose of advising a non-expert or assisting an expert. The model on which the reasoning is based can include high volumes of facts, crisp logic rules (PROLOG-type inference) with substantial add-ons for handling statistical and other operations on sets of facts, as well as inexact reasoning rules for handling certainties and probabilities (certainty rules) and fuzzy set rules (fuzzy set membership functions) for handling inexact classifications. |
|
|
The examples in the What is
Pro/3-web is mostly from the "MC" knowledge base* - a small demo application
which can help a user selecting a suitable motorbike. It can be downloaded from this site.
The diagram over shows dependencies between different
"knowledge entities" in the MC KB, that is, all knowledge
entities which lead up to the concluded sentence (set) manufacturer has
suitable bike.
* Note that some examples are from older versions of this model. |
|
 |
|
