CaseBasedReasoning (version 0.1)

RFModel: RandomForest Proximity

Description

This class uses the proximity matrix of the random survival forest algorithm as a similarity matrix (sqrt(1 - proximity matrix)) of learning and verum cases. By default all cases with at least one missing values are dropped from learning, calculating the distance matrix, and searching for similar cases.

Usage

RFModel

Arguments

Format

R6Class object

Usage

For usage details see Methods, Arguments, and Examples sections.

Methods

new(formula, ...)

This method is used to create an object of this class RFModel. Formula for analysis has to be provided. Further parameters for the ranger call can be provided here.

fit(dtData)

Fits the RandomForest model.

set_dist(method)

Set the proximity measure, 'Depth' (Default) and 'Proximity' are allowed.

...

See CBRBase class.

References

Englund and Verikas. A novel approach to estimate proximity in a random forest: An exploratory study.