minimax: Return sorted data points using minimax algorithm

In lsheremet/imbc:

Description

Return sorted data points using minimax algorithm

Usage

minimax(data, estep, previousResultIndicies, distanceMethod = "euclidean")

Arguments

data	the data you wish to use
estep	output from the mclust function "estep"
previousResultIndicies	indicies of rows the imbc algorithm previously queried the user on
distanceMethod	a method used to find distances between points. This must be one of "euclidean", "maximum", "manhattan", "canberra", "binary" or "minkowski", with a default of "euclidean."

Value

a vector of points sorted according to the minimax algorithm

Examples

#Load data
library(mclust)
data(banknote)

#Create new dataset with only continuous variables
bankdata <- banknote[,2:7]

#create Mclust object
object <- Mclust(bankdata)
#determine best model
model <- object$modelName
#extract model parameters
param <- object$parameters

#output from estep of EM algorithm
estepbank <- estep(model, bankdata, param)

#we assume we have not previously queried any points
previousResultIndicies <- c()

#output from minimax algorithm: returns vector of points that are least confidently placed in their respective classes
minimax(bankdata, estep, previousResultIndicies)

#try with a different distance method
minimax(bankdata, estep, previousResultIndicies, distanceMethod = "manhattan")

lsheremet/imbc documentation built on May 20, 2019, 7:01 p.m.