optional tree

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

optional tree

Ankie
Hi all,
 
How do I interpret an optional tree?
So when I have five nodes to choose from in the first step, how does it choose?
 
If it is different for each algorith, then I want to know it for the ADtree (alternating decision tree).
 
Thanks,
Ankie

_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
Reply | Threaded
Open this post in threaded view
|

Re: optional tree

Eibe Frank
You have to follow all possible paths at the option nodes (the
rectangles) and one of the paths at each decision node (the ellipses).
Sum up all the numbers that you encounter in the process. If the result
is positive, you predict one class, otherwise the other class.

Cheers,
Eibe

On Jul 6, 2005, at 1:30 AM, Ankie wrote:

> Hi all,
>  
> How do I interpret an optional tree?
> So when I have five nodes to choose from in the first step, how does
> it choose?
>  
> If it is different for each algorith, then I want to know it for the
> ADtree (alternating decision tree).
>  
> Thanks,
> Ankie_______________________________________________
> Wekalist mailing list
> [hidden email]
> https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist


_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
Reply | Threaded
Open this post in threaded view
|

Re: optional tree

Ankie
But which path should I follow at the first decision node. The upper elipse
has five branches.
So how do I know which one to take?


----- Original Message -----
From: "Eibe Frank" <[hidden email]>
To: "Ankie" <[hidden email]>
Cc: <[hidden email]>
Sent: Tuesday, July 05, 2005 11:11 PM
Subject: Re: [Wekalist] optional tree


You have to follow all possible paths at the option nodes (the
rectangles) and one of the paths at each decision node (the ellipses).
Sum up all the numbers that you encounter in the process. If the result
is positive, you predict one class, otherwise the other class.

Cheers,
Eibe

On Jul 6, 2005, at 1:30 AM, Ankie wrote:

> Hi all,
>
> How do I interpret an optional tree?
> So when I have five nodes to choose from in the first step, how does it
> choose?
>
> If it is different for each algorith, then I want to know it for the
> ADtree (alternating decision tree).
>
> Thanks,
> Ankie_______________________________________________
> Wekalist mailing list
> [hidden email]
> https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist



_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist