You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@mahout.apache.org by venkata ramana <ve...@gmail.com> on 2014/02/13 06:38:53 UTC

Integrate two Navie Bayes models

Hi All,

I have developed two mahout naive bayes models. Two models are built based
on the similar data. Let suppose I have 10 categories data. I split the
data where row count greater than 100000 as one first model input and
remaining is second model input.

Both are working fine. Now I would like to integrate/combine those models
and test it at glance.

Please let me know if any suggestions.

Thanks,
Venkat

Re: Integrate two Navie Bayes models

Posted by Ted Dunning <te...@gmail.com>.
Just keep both models and evaluate each one.


On Thu, Feb 13, 2014 at 1:16 AM, venkata ramana <venkat.ecosystems@gmail.com
> wrote:

> Hi,
>
> These models I am using for on the fly URL categorization. I can deploy the
> models seperately on webserver, also I am able to classify the URL and
> category. Do we have any option to put both models at a time?  My
> expectation is classify the url even if it falls on model one or two.
>
> Thanks in advance.
>
> Reg,
> Venkat
>
>
> On Thu, Feb 13, 2014 at 1:16 PM, Ted Dunning <te...@gmail.com>
> wrote:
>
> > What do you mean by "combine"?
> >
> >
> >
> >
> > On Wed, Feb 12, 2014 at 9:38 PM, venkata ramana <
> > venkat.ecosystems@gmail.com
> > > wrote:
> >
> > > Hi All,
> > >
> > > I have developed two mahout naive bayes models. Two models are built
> > based
> > > on the similar data. Let suppose I have 10 categories data. I split the
> > > data where row count greater than 100000 as one first model input and
> > > remaining is second model input.
> > >
> > > Both are working fine. Now I would like to integrate/combine those
> models
> > > and test it at glance.
> > >
> > > Please let me know if any suggestions.
> > >
> > > Thanks,
> > > Venkat
> > >
> >
>

Re: Integrate two Navie Bayes models

Posted by venkata ramana <ve...@gmail.com>.
Hi,

These models I am using for on the fly URL categorization. I can deploy the
models seperately on webserver, also I am able to classify the URL and
category. Do we have any option to put both models at a time?  My
expectation is classify the url even if it falls on model one or two.

Thanks in advance.

Reg,
Venkat


On Thu, Feb 13, 2014 at 1:16 PM, Ted Dunning <te...@gmail.com> wrote:

> What do you mean by "combine"?
>
>
>
>
> On Wed, Feb 12, 2014 at 9:38 PM, venkata ramana <
> venkat.ecosystems@gmail.com
> > wrote:
>
> > Hi All,
> >
> > I have developed two mahout naive bayes models. Two models are built
> based
> > on the similar data. Let suppose I have 10 categories data. I split the
> > data where row count greater than 100000 as one first model input and
> > remaining is second model input.
> >
> > Both are working fine. Now I would like to integrate/combine those models
> > and test it at glance.
> >
> > Please let me know if any suggestions.
> >
> > Thanks,
> > Venkat
> >
>

Re: Integrate two Navie Bayes models

Posted by Ted Dunning <te...@gmail.com>.
What do you mean by "combine"?




On Wed, Feb 12, 2014 at 9:38 PM, venkata ramana <venkat.ecosystems@gmail.com
> wrote:

> Hi All,
>
> I have developed two mahout naive bayes models. Two models are built based
> on the similar data. Let suppose I have 10 categories data. I split the
> data where row count greater than 100000 as one first model input and
> remaining is second model input.
>
> Both are working fine. Now I would like to integrate/combine those models
> and test it at glance.
>
> Please let me know if any suggestions.
>
> Thanks,
> Venkat
>