Posted by
Marcus G. Daniels on
Feb 20, 2015; 6:42pm
URL: http://friam.383.s1.nabble.com/Fwd-Share-Your-Knowledge-Taxonomy-Boot-Camp-tp7586090p7586091.html
On Fri, 2015-02-20 at 09:26 -0700, Tom Johnson wrote:
> Interesting topic here, at least to me. Has anyone ever attended
> this?
Have not. Some folks, like catalogers & librarians are good at this
sort of thing, it seems very tedious and hard to scale.
To scale it, I would imagine post-processing models built from
statistical inference techniques. One could imagine having `cold' models
which were heavily used and tested in predictive contexts, but also
`hot' models that were closer to the raw ingestion of the information --
vague correlations like one might establish after attending a talk on a
unfamiliar subject or interpreting Google results.
I've always thought it would be interesting to see if, say, graphs from
the latter type could be hardened using logical reasoning (and a corpus
of reliable models) into the `cold' type. A company I always associate
with this is Cycorp. I see they have an open position along these
lines!
http://www.cyc.com/careers/senior-ontologistObviously, there's IBM Watson that would have explored some of this
territory, and probably more on the medical apps side. Not sure how
much of Watson in practice comes from pure reasoning. Winning on
Jeopardy is one thing, but consuming analytical or technical literature
to do accurate Q&A is another.
Marcus
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe
http://redfish.com/mailman/listinfo/friam_redfish.com