[prev in list] [next in list] [prev in thread] [next in thread] 

List:       wekalist
Subject:    Re: [Wekalist] Empty instances
From:       Eibe Frank <eibe () waikato ! ac ! nz>
Date:       2018-05-17 1:50:59
Message-ID: D03B9BB4-2272-4649-B1D3-D9C420DF9F38 () waikato ! ac ! nz
[Download RAW message or body]

I suppose there are two different types of "empty" instances: ones where all \
predictor attributes have value zero, and ones where all predictor attribute values \
are missing. You are probably referring to the former case because you are working on \
text classification. In either case, treatment depends on the learning algorithm that \
is applied.

If all predictor attributes have value zero, and you are using a linear classifier \
such as a linear SVM, then classification will be based on the intercept only.

Cheers,
Eibe

> On 17/05/2018, at 6:04 AM, joncarv <joncarv@gmail.com> wrote:
> 
> Hi All,
> 
> Does anyone know how Weka classifies empty test instances?
> 
> I'm applying feature selection in a text classification problem, but some
> documents (instances) cannot be represented by the set of selected features
> and they are represented as an empty vector.
> I am not sure how Weka assigns a class for those instances...
> 
> Thanks a lot,
> Jonnathan.
> 
> 
> 
> --
> Sent from: http://weka.8497.n7.nabble.com/
> _______________________________________________
> Wekalist mailing list
> Send posts to: Wekalist@list.waikato.ac.nz
> List info and subscription status: \
> https://list.waikato.ac.nz/mailman/listinfo/wekalist List etiquette: \
> http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list
Send posts to: Wekalist@list.waikato.ac.nz
List info and subscription status: \
https://list.waikato.ac.nz/mailman/listinfo/wekalist List etiquette: \
http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html


[prev in list] [next in list] [prev in thread] [next in thread]