Learn R Programming

KnowGRRF (version 1.0)

Knowledge-Based Guided Regularized Random Forest

Description

Random Forest (RF) and Regularized Random Forest can be used for feature selection. Moreover, by Guided Regularized Random Forest, statistical-based weights are used to guide the regularization of random forest and further used for feature selection. This package can integrate prior information from multiple domains (statistical based and knowledge domain) to guide the regularization of random forest and feature selection. For more details, see reference: Guan X., Liu L. (2018) .

Copy Link

Version

Install

install.packages('KnowGRRF')

Monthly Downloads

2

Version

1.0

License

GPL (>= 2)

Maintainer

Xin Guan

Last Published

March 6th, 2019

Functions in KnowGRRF (1.0)

rrf.opt.m

KnowGRRF with weights from multiple knowledge domain
select.stable

Select a set of stable features based on frequency picked by GRRF.
get.performance

Get performance of feature selection
on.aic

AIC from model built with KnowGRRF, functions used in optimization to find scaling parameter for rrf.opt.1 or rrf.opt.m
rf.repeat

Build random forest multiple times and return AUC for both training and test set if available
rrf.once

Feature selection by regularized random forest and compare against full model
select.stable.aic

Select a set of stable features based on AIC after an initial selection by GRRF
rf.once

Build random forest once and return AUC for both training and test set if available
write.roc

write test ROC to a data table.
rrf.opt.1

KnowGRRF with weights from one knowledge domain