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-ontologist Obviously, 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
