Ideas from Wrappers for Feature Subset Selection


  • Paper is as survey of feature selection algorithms, was written in ’96 so its probably dated by now.
  • Discussion of various definitions of relevance/irrelevance.
  • Demonstration that in imperfect learners (as opposed to bayes optimal) relevance of a feature doesn’t necessitate that it is in the optimal feature subset, conversely that irrelevance doesn’t imply it shouldn’t be in the optimal feature subset.  These are basically due to restrictions in the hypothesis space real algorithms have.
Advertisements
Tagged , ,

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: