Quantcast
Channel: Pentaho Community Forums
Viewing all articles
Browse latest Browse all 16689

Ignore a feature in classification but include in prediction output w/ GUI or R?

$
0
0
Hello,

I am trying to retain a couple of features (date and ID) in my data file that I would like a supervised learner to ignore during classification, but output during prediction. I understand from searching for answers to this on Stack Overflow that you can used the FilteredClassifier meta-filter when using Weka in Java, but I'm wondering if there is a way to do this in either the GUI or using the RWeka package in my R code.

Note: I did try to use the "addID" filter in the GUI but I found that, as I suspected, it negatively affected classification accuracy.

Thank you,

Nick

Viewing all articles
Browse latest Browse all 16689

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>