HELP! Weka: Prepocess - From Categorical (ordinal) to numeric

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

HELP! Weka: Prepocess - From Categorical (ordinal) to numeric

Chris Jensen
Hi,

Im doing a predictive analysis of the following dataset:
https://www.kaggle.com/shivam2503/diamonds

The dataset includes both numeric and categorical values. I want to do a
predictive model using regression in weka. The problem is that i dont know
if it will be possible when there are both numeric and categorical values.
Im thinking that i should converte the to numeric but om not sure how.

Hope you guys can help!




--
Sent from: https://weka.8497.n7.nabble.com/
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: 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: HELP! Weka: Prepocess - From Categorical (ordinal) to numeric

Eibe Frank
The OrdinalToNumeric filter provides a very basic way to turn nominal attributes with an ordered set of values into numeric ones. The nominal values are mapped to their position in the list of values given in the definition of the attribute.

The resulting attribute is clearly not normally interval or ratio scale (unless the ordinal values can be considered equi-distant, in which case the interval scale assumption is reasonable). However, there are learners such as decision tree and rule learners like J48, RandomForest, and JRip that effectively treat every numeric attribute as a purely ordinal attribute anyway (where only the order of the attribute values matters and the distance between values is irrelevant) so this does not matter when you apply one of those learning algorithms.

Cheers,
Eibe

On Fri, Dec 6, 2019 at 11:13 AM Chris Jensen <[hidden email]> wrote:
Hi,

Im doing a predictive analysis of the following dataset:
https://www.kaggle.com/shivam2503/diamonds

The dataset includes both numeric and categorical values. I want to do a
predictive model using regression in weka. The problem is that i dont know
if it will be possible when there are both numeric and categorical values.
Im thinking that i should converte the to numeric but om not sure how.

Hope you guys can help!




--
Sent from: https://weka.8497.n7.nabble.com/
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: 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: 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