Re: Inconsistent models, mapping, interoperability, and the SUO

["Rich Cooper" <richcooper@mindspring.com>]

Rob Freeman wroteL
> On Friday 25 March 2005 00:46, Rich Cooper wrote:
>>
>>   As you can see, WordNet 2.0 models the English usage for drug naming
>> and
>> classification conventions, so it provides an initial charge of words,
>> relationships among words, and knowledge about drugs that helps reach a
>> fully capable system faster than if you just start with the words alone.
>
> I don't really see that. What I see is a dictionary. There is not much
> structure there. But I am guessing that if I open up the node labelled
> "antisyphilitic" I will see a list of antisyphilitics, and so on.
>
> Most of the information you need is not captured in those (assumed)
> hierarchical relationships though, is it? For instance the deadly
> "three-way
> pharmacodynamic drug-drug interaction" pictured in your first gif. How do
> you
> find that in WordNet?
>
> -Rob

Correct, that information is not in WordNet 2.0 and also not all of the
drugs that have to be watched for are listed in WordNet either.

The point is that WordNet contains the initial charge of common
English terms, a lattice of nouns, the synsets of nouns, verbs, adjectives
and adverbs in common usage.  It also contains a number of phrases,
definition glosses, and other useful material.

To apply this generic English to specific applications, such as drug
interactions, means there must be more information supplied which is
dependent on the application.  Pharmaceutical directories of drugs,
for example, and their known interactions.

So WordNet 2.0 is useful, but not completely sufficient, for these
purposes.

The hard part though, is not the words and phrases, but the meaning
they have and how to infer useful alerts from the collection of
text given a collection of words.  That's where the deductive part
comes in.  There's a lot more to it than just the general slides I showed
in the earlier email.

Rich