SelectResult-class: Container for Storing Feature Selection Results

In ClassifyR: A framework for cross-validated classification problems, with applications to differential variability and differential distribution testing

Description

Contains a list of ranked names of features, from most discriminative to least discriminative, and a list of features selected for use in classification. The names will be in a data frame if the input data set is a MultiAssayExperiment, with the first column containing the name of the data table the feature is from and the second column containing the name of the feature. Each vector or data frame element in the list corresponds to a particular iteration of classifier training. Nested lists will be present if the permutation and folding cross-validation scheme was used. This class is not intended to be created by the user, but could be used in another software package.

Constructor

SelectResult(datasetName, selectionName, totalFeatures, rankedFeatures, chosenFeatures)

datasetName

A name associated with the data set used.

selectionName

A name associated with the classification.

totalFeatures

The total number of features in the data set.

rankedFeatures

Identifiers of all features or meta-features if meta-features were used by the classifier, from most to least discriminative.

chosenFeatures

Identifiers of features or meta-features if meta-features were used by the classifier selected at each fold.

Summary

A method which summarises the results is available. result is a SelectResult object.

show(result)Prints a short summary of what result contains.

Author(s)

Dario Strbenac

Examples

    SelectResult("Asthma", "Moderated t-test", 50, list(1:50), list(1:10))

ClassifyR documentation built on Nov. 8, 2020, 6:53 p.m.