Hitachi Vantara Pentaho Community Wiki
Child pages
  • LinearForwardSelection
Skip to end of metadata
Go to start of metadata

Package

weka.attributeSelection

Synopsis

LinearForwardSelection:

Extension of BestFirst. Takes a restricted number of k attributes into account. Fixed-set selects a fixed number k of attributes, whereas k is increased in each step when fixed-width is selected. The search uses either the initial ordering to select the top k attributes, or performs a ranking (with the same evalutator the search uses later on). The search direction can be forward, or floating forward selection (with opitional backward search steps).

For more information see:

Martin Guetlein (2006). Large Scale Attribute Selection Using Wrappers. Freiburg, Germany.

Available in Weka 3.6.x - 3.7.1. Available via the package management system for Weka >= 3.7.2 (linearForwardSelection).

Options

The table below describes the options available for LinearForwardSelection.

Option

Description

forwardSelectionMethod

Set the direction of the search.

lookupCacheSize

Set the maximum size of the lookup cache of evaluated subsets. This is expressed as a multiplier of the number of attributes in the data set. (default = 1).

numUsedAttributes

Set the amount of top-ranked attributes that are taken into account by the search process.

performRanking

Perform initial ranking to select top-ranked attributes.

searchTermination

Set the amount of backtracking. Specify the number of

startSet

Set the start point for the search. This is specified as a comma seperated list off attribute indexes starting at 1. It can include ranges. Eg. 1,2,5-9,17.

type

Set the type of the search.

verbose

Turn on verbose output for monitoring the search's progress.

  • No labels