filter_features_AUC: A function for filtering-based feature selection, using...

In skimlab/CCSBUtils: R Utilitiy Functions for CCSB

A function for filtering-based feature selection, using ROC/AUC as a metric

Description

A function for filtering-based feature selection, using ROC/AUC as a metric

Usage

filter_features_AUC(
  dd,
  cls,
  threshold = 0.8,
  direction = c("two.sided", "greater", "less")
)

Arguments

dd	A matrix or data.frame where each row is an observation
cls	A set of classes, either in factor
threshold	A minimum value of AUC for filtering
direction	Direction of AUC statistical test: "two.sided", "greater" or "less"

Value

A data.frame of feature, diff, score where score > threshold

skimlab/CCSBUtils documentation built on March 30, 2022, 4:52 a.m.