Quantcast

ClassificationViaClustering

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

ClassificationViaClustering

valerio jus

Hi Weka list,

Can onyone explain in details the mechanism of Weka's "ClassificationViaClustering" when having a dataset with 5 class labels. In other words, how each of the five classes are going to evaluate the resulting clusters of the base clusterer method like EM?

Thanks.
Valerio


_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
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
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Fwd: ClassificationViaClustering

valerio jus

Hi Weka list,

Can onyone explain in details the mechanism of Weka's "ClassificationViaClustering" when having a dataset with 5 class labels. In other words, how each of the five classes are going to evaluate the resulting clusters of the base clusterer method like EM?

Thanks.
Valerio

---------- Forwarded message ----------
From: "valerio jus" <[hidden email]>
Date: 9 Jan 2017 6:22 pm
Subject: ClassificationViaClustering
To: "Weka machine learning workbench list." <[hidden email]>
Cc:

Hi Weka list,

Can onyone explain in details the mechanism of Weka's "ClassificationViaClustering" when having a dataset with 5 class labels. In other words, how each of the five classes are going to evaluate the resulting clusters of the base clusterer method like EM?

Thanks.
Valerio


_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
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
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: ClassificationViaClustering

Eibe Frank-2
Administrator
Here is an explanation of how WEKA maps classes to a set of clusters that have been found without considering the class attribute:

  http://weka.8497.n7.nabble.com/classes-to-clusters-td6684.html#a6685

This process is used in both the "classes-to-clusters evaluation” in the clustering panel and the ClassificationViaClustering method for classification.

Cheers,
Eibe

> On 12/01/2017, at 2:36 PM, valerio jus <[hidden email]> wrote:
>
> Hi Weka list,
>
> Can onyone explain in details the mechanism of Weka's "ClassificationViaClustering" when having a dataset with 5 class labels. In other words, how each of the five classes are going to evaluate the resulting clusters of the base clusterer method like EM?
>
> Thanks.
> Valerio
>
> ---------- Forwarded message ----------
> From: "valerio jus" <[hidden email]>
> Date: 9 Jan 2017 6:22 pm
> Subject: ClassificationViaClustering
> To: "Weka machine learning workbench list." <[hidden email]>
> Cc:
>
> Hi Weka list,
>
> Can onyone explain in details the mechanism of Weka's "ClassificationViaClustering" when having a dataset with 5 class labels. In other words, how each of the five classes are going to evaluate the resulting clusters of the base clusterer method like EM?
>
> Thanks.
> Valerio
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> 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: [hidden email]
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
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: ClassificationViaClustering

valerio jus

Thanks a lot,  Eibe.

Valerio

On 12 Jan 2017 10:09 am, "Eibe Frank" <[hidden email]> wrote:
Here is an explanation of how WEKA maps classes to a set of clusters that have been found without considering the class attribute:

  http://weka.8497.n7.nabble.com/classes-to-clusters-td6684.html#a6685

This process is used in both the "classes-to-clusters evaluation” in the clustering panel and the ClassificationViaClustering method for classification.

Cheers,
Eibe

> On 12/01/2017, at 2:36 PM, valerio jus <[hidden email]> wrote:
>
> Hi Weka list,
>
> Can onyone explain in details the mechanism of Weka's "ClassificationViaClustering" when having a dataset with 5 class labels. In other words, how each of the five classes are going to evaluate the resulting clusters of the base clusterer method like EM?
>
> Thanks.
> Valerio
>
> ---------- Forwarded message ----------
> From: "valerio jus" <[hidden email]>
> Date: 9 Jan 2017 6:22 pm
> Subject: ClassificationViaClustering
> To: "Weka machine learning workbench list." <[hidden email]>
> Cc:
>
> Hi Weka list,
>
> Can onyone explain in details the mechanism of Weka's "ClassificationViaClustering" when having a dataset with 5 class labels. In other words, how each of the five classes are going to evaluate the resulting clusters of the base clusterer method like EM?
>
> Thanks.
> Valerio
>
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> 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: [hidden email]
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: [hidden email]
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
Loading...