Re: [bpfk] CLL errata check: lujvo scoring

[Robin Lee Powell <rlpowell@digitalkingdom.org>]

On Wed, Oct 27, 2010 at 04:48:52PM +0200, Arnt Richard Johansen
wrote:
> On Wed, Oct 27, 2010 at 11:03:50AM -0300, Jorge Llambías wrote:
> > The algorithm doesn't take consonant clusters into account, so
> > I'm not sure how often it gives a choice that people find
> > preferable. It prefers "kixsku" to "ki'asku" for example,

*UGH*.

> It might be fun and instructive for some of the computer guys to
> come up with a machine learning algorithm that optimises the lujvo
> scoring algorithm to reflect what Lojbanists actually like.
> (Perhaps using published texts as a basis rather than asking them
> to choose between different versions of the same lujvo.)

I don't think there's enough of the right kind of corpus for that; I
think everybody just did what jvocuhadju said.  :)  I know I did.  I
never hit a case like kixsku.

Regardless, fixing the algorithm (while it sure sounds like a good
idea to me) is not relevant to my *current* enterprise.  :)

-Robin

-- 
http://singinst.org/ :  Our last, best hope for a fantastic future.
Lojban (http://www.lojban.org/): The language in which "this parrot
is dead" is "ti poi spitaki cu morsi", but "this sentence is false"
is "na nei".   My personal page: http://www.digitalkingdom.org/rlp/

-- 
You received this message because you are subscribed to the Google Groups "BPFK" group.
To post to this group, send email to bpfk-list@googlegroups.com.
To unsubscribe from this group, send email to bpfk-list+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/bpfk-list?hl=en.