mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "j.barrett Strausser" <j.barrett.straus...@gmail.com>
Subject Re: Mahout 0.8 Random Forest Accuracy
Date Fri, 18 Oct 2013 14:50:04 GMT
Just a theoretical note : Accuracy isn't the best metric for a Classifier.
Mahouts accuracy could well be less than the comparable R result, but be a
better classifier, at least according to the F1 metric.

How difficult would it be to wrap the RF classifier into an ensemble
learner?

-barrett


On Fri, Oct 18, 2013 at 10:42 AM, Ted Dunning <ted.dunning@gmail.com> wrote:

> On Fri, Oct 18, 2013 at 7:48 AM, Tim Peut <tim@timpeut.com> wrote:
>
> > Has anyone found that Mahout's random forest doesn't perform as well as
> > other implementations? If not, is there any reason why it wouldn't
> perform
> > as well?
> >
>
> This is disappointing, but not entirely surprising.  There has been
> considerably less effort applied to Mahouts random forest package than the
> comparable R packages.
>
> Note, particularly that the Mahout implementation is not regularized.  That
> could well be a big difference.
>



-- 


https://github.com/bearrito
@deepbearrito

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message