You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@spark.apache.org by Aditya Addepalli <dy...@gmail.com> on 2020/12/02 13:30:13 UTC

Remove subsets from FP Growth output

Hi,

Is there a good way to remove all the subsets of patterns from the output
given by FP Growth?

For example if both the patterns pass the confidence and support thresholds:

[Attribute1 = A, Attribute2 = B] -> [Output=C]
[Attribute1 = A] -> [Output=C]

I want to choose only [Attribute1 = A] -> [Output=C]

Any ideas that come to mind would be appreciated.

Re: Remove subsets from FP Growth output

Posted by Sean Owen <sr...@gmail.com>.

-dev

Increase the threshold? Just filter the rules as desired after they are
generated?
It's not clear what your criteria are.

On Wed, Dec 2, 2020 at 7:30 AM Aditya Addepalli <dy...@gmail.com> wrote:

> Hi,
>
> Is there a good way to remove all the subsets of patterns from the output
> given by FP Growth?
>
> For example if both the patterns pass the confidence and support
> thresholds:
>
> [Attribute1 = A, Attribute2 = B] -> [Output=C]
> [Attribute1 = A] -> [Output=C]
>
> I want to choose only [Attribute1 = A] -> [Output=C]
>
> Any ideas that come to mind would be appreciated.
>