AttributeSelectedClassifier

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

AttributeSelectedClassifier

asadbtk
Hello Eibe and Peter

This diagram below uses the AttributeSelectedClassifier architecture for feature selection and learning algorithms. I could not understand the last part of (A). In theory, the authors mentioned that " the reduced training set is used to infer a classifier f().

What does this mean?

The diagram is taken from this paper:

Best regards



image.png

_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to [hidden email]
To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|

Re: AttributeSelectedClassifier

Peter Reutemann
> This diagram below uses the AttributeSelectedClassifier architecture for feature selection and learning algorithms. I could not understand the last part of (A). In theory, the authors mentioned that " the reduced training set is used to infer a classifier f()."
>
> What does this mean?

I'm not sure what's not to understand...

There are two phases for a batch classifier: train and predict.
- train
The AttributeSelectedClassifier receives training data, pushes it
through the feature selection setup that you defined to obtain a
reduced set of attributes to use.
This reduced dataset is used to build the specified base classifier.
- predict
The AttributeSelectedClassifier extracts the subset of attribute
values that were determined during training and presents these to the
base classifier model for making a prediction and returns that.

"to infer a classifier f()" is just a fancy way of saying "training
the base classifier on the attribute subset" (any classifier can be
viewed as a function from a mathematical point of view).

Cheers, Peter
--
Peter Reutemann
Dept. of Computer Science
University of Waikato, NZ
+64 (7) 858-5174
http://www.cms.waikato.ac.nz/~fracpete/
http://www.data-mining.co.nz/
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to [hidden email]
To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|

Re: AttributeSelectedClassifier

asadbtk
Thanks a lot Peter for the clarification.

Best regards

On Sun, May 3, 2020 at 1:31 AM Peter Reutemann <[hidden email]> wrote:
> This diagram below uses the AttributeSelectedClassifier architecture for feature selection and learning algorithms. I could not understand the last part of (A). In theory, the authors mentioned that " the reduced training set is used to infer a classifier f()."
>
> What does this mean?

I'm not sure what's not to understand...

There are two phases for a batch classifier: train and predict.
- train
The AttributeSelectedClassifier receives training data, pushes it
through the feature selection setup that you defined to obtain a
reduced set of attributes to use.
This reduced dataset is used to build the specified base classifier.
- predict
The AttributeSelectedClassifier extracts the subset of attribute
values that were determined during training and presents these to the
base classifier model for making a prediction and returns that.

"to infer a classifier f()" is just a fancy way of saying "training
the base classifier on the attribute subset" (any classifier can be
viewed as a function from a mathematical point of view).

Cheers, Peter
--
Peter Reutemann
Dept. of Computer Science
University of Waikato, NZ
+64 (7) 858-5174
http://www.cms.waikato.ac.nz/~fracpete/
http://www.data-mining.co.nz/
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to [hidden email]
To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to [hidden email]
To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html