[Corpora-List] The Implement of LDA classifier

[Jason Eisner]

On Thu, Jul 29, 2010 at 8:23 AM, CARLOS EDUARDO DANTAS DE MENEZES <
cedmenezes at gmail.com> wrote:
> I've found an e-mail to corpora list that seems to answer your question.

Except that since the question was about classifiers, "LDA" probably
referred to Linear Discriminant Analysis, not Latent Dirichlet Allocation.
:-)

(But are implementations really so hard to find?  Just search for linear
discriminant analysis (or fisher discriminant). The top Google hit for this
search is the wikipedia
page<http://en.wikipedia.org/wiki/Linear_discriminant_analysis>,
which links to some implementations at the bottom including
C++/etc.<http://www.alglib.net/dataanalysis/lineardiscriminantanalysis.php>.
 Or if you want a Matlab
implementation <http://www.mathworks.com/matlabcentral/fileexchange/189>, R
implementation<http://stat.ethz.ch/R-manual/R-devel/library/MASS/html/lda.html>,
etc., then add "matlab", "R", etc. to your query.  Note: I am not
specifically recommending any of these implementations ...)

On Tue, Jul 27, 2010 at 9:40 PM, Xiao-lin Wang <arthur_general at sjtu.edu.cn>
wrote:
> My research topic at present is meta-learning. I have been trying SVM as
the
> meta-level classifier, and there are several popular SVM implements on the
> web. But SVM turns out to work poor mainly due to the computational
> complexity, which costs more than several weeks to learn a large
meta-level
> training set.   Recently I want to try LDA classifier as the meta-level
> classifier, but I can not find a implement of LDA classifiers after
spending
> hours on the web. Can any body recommend some implements of LDA?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20100729/759abd96/attachment.htm>
-------------- next part --------------
_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora