I received an offline note with a few questions about this issue.
Since this topic is related to my slides from a panel discussion
last Thursday, I'll suggest them for more info: (01)
http://www.jfsowa.com/talks/soup_llr.pdf (02)
> I'd like to know more about how you used WordNet
> in your work at VivoMind. (03)
Short answer: As a lexical resource for use in parsing. We also
use Roget's Thesaurus in the same way as WordNet. Neither one has
a dependable hierarchy, but they're OK as a rough approximation.
Roget's Th. is much better than WN for adjectives and adverbs.
They supplement one another. But I would *not* recommend an
attempt to "harmonize" them in a single resource. (04)
> My present thinking is that synsets are "fuzzy" concepts
> which can be used for higher-level reasoning, which
> can guide further thinking and actions. (05)
I would say that the hierarchy of WordNet (or Roget's) is better
than no hierarchy at all. But the ultimate solution is to derive
a distinct ontology for each document or set of documents. A rough
first pass, such as WN or RT, is useful to start with. (06)
> It sounds like concept-graph-matching provides
> a "ContextNet" for your VivoMind work. (07)
The ontology for any single document is usually fairly consistent.
Graphs derived from simple sentences in one part of the document
can be used as a basis for interpreting other sentences that may
be less clear. (08)
John (09)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: 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 join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (010)
|