Cross-validation Folds 10

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Cross-validation Folds 10

ji an
Hi All,

could anyone answer my question about
cross-validation?

if Folds is set to 10, we can get 10 different
training datasets. this means we will get 10 different
rules groups, but only one
rule group was extracted in my experiment. (I tested
decision tree J48). could you tell me why?

Thanks

Anj


               
____________________________________________________
Yahoo! Sports
Rekindle the Rivalries. Sign up for Fantasy Football
http://football.fantasysports.yahoo.com

_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
Reply | Threaded
Open this post in threaded view
|

Re: Cross-validation Folds 10

Lei Tang
The evaluation result is based on 10-fold cross validation.
But the tree given is built based on all the training data.

--Lei

On 6/23/05, ji an <[hidden email]> wrote:
Hi All,

could anyone answer my question about
cross-validation?

if Folds is set to 10, we can get 10 different
training datasets. this means we will get 10 different
rules groups, but only one
rule group was extracted in my experiment. (I tested
decision tree J48). could you tell me why?

Thanks

Anj



____________________________________________________
Yahoo! Sports
Rekindle the Rivalries. Sign up for Fantasy Football
http://football.fantasysports.yahoo.com

_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist


_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
Reply | Threaded
Open this post in threaded view
|

Re: Cross-validation Folds 10

ji an
Hi all,
I think the train dataset should not include test
data.
do you know how I can extract 10 rule groups from
different 90% data sets for
cross validation folds 10?

Thanks Lei for your prompt reply.

ANJ


--- Lei Tang <[hidden email]> wrote:

> The evaluation result is based on 10-fold cross
> validation.
> But the tree given is built based on all the
> training data.
>
> --Lei
>
> On 6/23/05, ji an <[hidden email]> wrote:
> >
> > Hi All,
> >
> > could anyone answer my question about
> > cross-validation?
> >
> > if Folds is set to 10, we can get 10 different
> > training datasets. this means we will get 10
> different
> > rules groups, but only one
> > rule group was extracted in my experiment. (I
> tested
> > decision tree J48). could you tell me why?
> >
> > Thanks
> >
> > Anj
> >
> >
> >
> >
> ____________________________________________________
> > Yahoo! Sports
> > Rekindle the Rivalries. Sign up for Fantasy
> Football
> > http://football.fantasysports.yahoo.com
> >
> > _______________________________________________
> > Wekalist mailing list
> > [hidden email]
> >
>
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
> >
>


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com 

_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
Reply | Threaded
Open this post in threaded view
|

Re: Cross-validation Folds 10

Eibe Frank
You can get that information in the KnowledgeFlow user interface but
not in the Explorer.

Cheers,
Eibe

On Jun 24, 2005, at 2:33 AM, ji an wrote:

> Hi all,
> I think the train dataset should not include test
> data.
> do you know how I can extract 10 rule groups from
> different 90% data sets for
> cross validation folds 10?
>
> Thanks Lei for your prompt reply.
>
> ANJ
>
>
> --- Lei Tang <[hidden email]> wrote:
>
>> The evaluation result is based on 10-fold cross
>> validation.
>> But the tree given is built based on all the
>> training data.
>>
>> --Lei
>>
>> On 6/23/05, ji an <[hidden email]> wrote:
>>>
>>> Hi All,
>>>
>>> could anyone answer my question about
>>> cross-validation?
>>>
>>> if Folds is set to 10, we can get 10 different
>>> training datasets. this means we will get 10
>> different
>>> rules groups, but only one
>>> rule group was extracted in my experiment. (I
>> tested
>>> decision tree J48). could you tell me why?
>>>
>>> Thanks
>>>
>>> Anj
>>>
>>>
>>>
>>>
>> ____________________________________________________
>>> Yahoo! Sports
>>> Rekindle the Rivalries. Sign up for Fantasy
>> Football
>>> http://football.fantasysports.yahoo.com
>>>
>>> _______________________________________________
>>> Wekalist mailing list
>>> [hidden email]
>>>
>>
> https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
>>>
>>
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
>
> _______________________________________________
> Wekalist mailing list
> [hidden email]
> https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist


_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist