featureSelection_basic: Feature Selection for Block Biclusters in Binary Data

In edcurry/GABi: Framework for Generalized Subspace Pattern Mining

Description

A feature selection function for the GABi biclustering framework, based on the definition of a bicluster as a block of consistently high values across a submatrix within a binary dataset.

Usage

featureSelection.basic(cols)

Arguments

cols	Numeric vector representing a subset of the columns from x across which this solution's bicluster pattern is defined.

Details

A fast feature selection function is vital to the GABi framework of biclustering. In GABi, the bicluster problem is reformulated around the fact that each subset of the columns across a dataset will have one _maximal_ subset of rows that fit a specified pattern, and the submatrix defined by this maximal subset of rows will be the most interesting observation involving that subset of columns. Makes use of fitnessArgs a list of parameters in the environment of execution of the biclustering function GABi. Notably, the element consistency is used to apply a stringency threshold for selecting features (i.e. only those with the proportion of high values across the subset of samples being greater than consistency)

Value

Numeric vector representing the features (i.e. rows) from dataset x representing the maximal bicluster for the solution encoded by chr.

Author(s)

Ed Curry [e-mail]

edcurry/GABi documentation built on May 16, 2019, 7:10 p.m.