train.cade: CADE

Description Usage Arguments

View source: R/cade.R

Description

R implementation of Classifier-Adjusted Density Estimation, as described by Friedland, Gentzel, and Jensen in their white paper "Classifier-Adjusted Density Estimation for Anomaly Detection and One-Class Classification"

Usage

1
2
train.cade(df, classifier = "randomForest",
  density.estimate = "uniform", true.prop = 0.5)

Arguments

df

data frame used for training

classifier

classifier used to distinguish between true and false data points, must be one of:

  • 'randomForest'

  • 'KNN'

  • 'logistic'

  • 'rpart'

  • 'nb'

density.estimate

method used to create false data points, must be one of

  • 'uniform'

  • 'Gaussian'

  • 'KDE'

true.prop

proportion of true data points in final training set


Prometheus77/rcade documentation built on Oct. 30, 2019, 10:44 p.m.