The Data Mining Forum 
IMPORTANT: This is the old Data Mining forum.
I keep it online so that you can read the old messages.
Please post your new messages in the new forum: https://forum2.philippe-fournier-viger.com/index.php
Adding confidence to sequential rule mining
Posted by:
Ron
Date: October 23, 2012 10:31AM
Hi All,
I'm wondering if I can get a mincof, or confidence metric by combining one or more of the algorithms that can calculate a confidence value with a rules algorithm that can't provide a confidence value?
For example, what could I do with the results of running two algorithms on the same sequence data? Could I combine the results somehow?
Regards,
Ron
I'm wondering if I can get a mincof, or confidence metric by combining one or more of the algorithms that can calculate a confidence value with a rules algorithm that can't provide a confidence value?
For example, what could I do with the results of running two algorithms on the same sequence data? Could I combine the results somehow?
Regards,
Ron
Re: Adding confidence to sequential rule mining
Posted by:
webmasterphilfv
Date: October 30, 2012 07:46AM
Hi Ron,
Sorry for the delay, I had a lot of things to do recently...
I think that there could be some utility to combine the output of some algorithms for some applications. It may depends on the application.
For the sequential rule mining algorithms in SPMF, they all use the minconf threshold already.
But they is a lot of possibility for improving the algorithms or making some variations of the algorithms.
I think that it depends on what you want to do with them!
Best,
Philippe
Sorry for the delay, I had a lot of things to do recently...
I think that there could be some utility to combine the output of some algorithms for some applications. It may depends on the application.
For the sequential rule mining algorithms in SPMF, they all use the minconf threshold already.
But they is a lot of possibility for improving the algorithms or making some variations of the algorithms.
I think that it depends on what you want to do with them!
Best,
Philippe