[Top] [All Lists]

Re: [ontolog-forum] Levels

To: "[ontolog-forum] " <ontolog-forum@xxxxxxxxxxxxxxxx>
From: Kathryn Blackmond Laskey <klaskey@xxxxxxx>
Date: Wed, 14 Feb 2007 21:32:32 -0500
Message-id: <p0611043cc1f971e0c76d@[]>
>KBL> Probabilistic information ought in some cases to be extremely
>  > high up in a "levels of entrenchment" ordering.
>The level of "entrenchment" has nothing to do with level of
>importance.  I would agree that a probability statement may be
>very important.  But an observation that "p is true" would
>overrule a statement that p has a 95% probability of being false.    (01)

The kind of probabilities that belong in an ontology won't be 
overridden in this way.  Instance-level probabilities don't go into 
an ontology.  The probabilities that go into an ontology apply to a 
class of individuals.    (02)

Base rates of conditions in defined populations belong in a medical 
ontology. Sensitivities and specificities of diagnostic tests belong 
in a medical ontology. Success rates for medical procedures performed 
under given conditions by persons with given qualifications belong in 
a medical ontology.  (Provided, of course, that we have good data to 
support the probabilistic assertions.)    (03)

The probability the physician assesses that Mary Smith has breast 
cancer prior to seeing the biopsy report for the lump he feels in her 
breast does not belong in a medical ontology.  Nor does the 
probability he assesses after feeling the lump.    (04)

After the biopsy definitively reveals cancerous cells in Mary's lump, 
the probability that Mary has breast cancer jumps from 5% to 100%. 
The assertion "Mary's lump contains cancerous cells" allows us to 
prove that "Mary has breast cancer" is true.  This overrides our 
prior assertion "There is a 95% chance that Mary does not have breast 
cancer."  However, the probabilities in the ontology (5% chance that 
someone with a lump in her breast has cancer; 100% chance that 
someone with a lump in her breast that contains cancerous cells has 
breast cancer) do not change.    (05)

Probabilities in ontologies will change as we learn more about the 
world.  But non-probabilistic content of ontologies also changes as 
we learn more about the world.    (06)

>I definitely agree.  But there may still be a few necessary statements,
>such as "every dog is an animal",    (07)

and "Every woman with a lump in her breast that contains cancerous 
cells has breast cancer."    (08)

>even though there may be no clear
>set of sufficient conditions for distinguishing a dog from a wolf or
>many other kinds of animals.    (09)

And just because the biopsy report says there are cancerous cells 
does not mean Mary has breast cancer.    (010)

Kathy    (011)

Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/  
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (012)

<Prev in Thread] Current Thread [Next in Thread>