The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here
[SOLVED] Output of Class Probability according to the Modell
Contributor I
Hi all,
I'm actually experimenting with classification on the German Data Set (CreditScoring), and was wondering if there is a possibility to get an output of the class probablities assigned to the Data by the classificator. I'm mainly working with the decision tree operator as it is easy to understand.
I also tried several other classificators but wasn't able to find something like this.
I hope this question isn't too stupid, because I'm relativly new to Data Mining and quite unsure about some of the things I'm acctually trying.
Edit: I think one of the confidence attributes added by the apply model operator might be what i'm looking for, but apparently I can't find any documantation for it. Also, if this is what I'm looking for, I'm not sure how to tell the decision tree operator where to set the cut-off, as it says the confidence parameter is only used for prepruning.
Edit2: I think i might have found what i was looking for. If I'm coorect, a can use the find and apply threshold operators. Am I correct thinking that the find threshold operator can also apply a costmatrix instead of the metacost operator?
Edit3: Ok, I'm done... I should have mentioned that i was working with ROC curves. What i was looking for was actually the create/apply threshold operators. Nonetheless, please tell me if I'm wrong in what I'm doing
Thanks in Advance
I'm actually experimenting with classification on the German Data Set (CreditScoring), and was wondering if there is a possibility to get an output of the class probablities assigned to the Data by the classificator. I'm mainly working with the decision tree operator as it is easy to understand.
I also tried several other classificators but wasn't able to find something like this.
I hope this question isn't too stupid, because I'm relativly new to Data Mining and quite unsure about some of the things I'm acctually trying.
Edit: I think one of the confidence attributes added by the apply model operator might be what i'm looking for, but apparently I can't find any documantation for it. Also, if this is what I'm looking for, I'm not sure how to tell the decision tree operator where to set the cut-off, as it says the confidence parameter is only used for prepruning.
Edit2: I think i might have found what i was looking for. If I'm coorect, a can use the find and apply threshold operators. Am I correct thinking that the find threshold operator can also apply a costmatrix instead of the metacost operator?
Edit3: Ok, I'm done... I should have mentioned that i was working with ROC curves. What i was looking for was actually the create/apply threshold operators. Nonetheless, please tell me if I'm wrong in what I'm doing
Thanks in Advance
0
