On Tue, Mar 16, 2010 at 10:02 AM, Marvin Humphrey
<[email protected]> wrote:
>
> Even very slow ones, eh?  How about one that requires gobs of RAM?

sure, i mean just thinking the algorithm could be prototyped first,
then made fast?

>
> This idea actually came out of a conversation I had with someone at an San
> Diego Ruby Users meeting who used to work on the OpenCyc classification
> engine.  From what I understand, the Cyc project is an AI project that sits on
> top of a kind of Yahoo directory or DMOZ for words.  Apparently it has a Java
> API and requires several GB of RAM to load.
>
>    http://www.cyc.com/cyc/opencyc
>    http://en.wikipedia.org/wiki/Cyc
>
> His suggestion was to use OpenCyc to classify terms.
>
> That's similar to what we'd do with topic vectors generated by an indexing
> component, except that the Cyc topic vectors were built laboriously by hand
> rather than using automatic dimension reduction.

at a quick glance, sounds possibly similar to
http://www.unc.edu/~jaguera/query-expansion/ (except it uses wordnet
instead)


-- 
Robert Muir
[email protected]

Reply via email to