scores.normalization: Scores Normalization Function

In gecko515/HEMDAG: Hierarchical Ensemble Methods for Directed Acyclic Graphs

Description

Functions to normalize a flat scores matrix w.r.t. max normalization (MaxNorm) or quantile normalization (Qnorm)

Usage

scores.normalization(norm.type = "MaxNorm", S)

Arguments

norm.type	can be one of the following two values: MaxNorm (def.): each score is divided w.r.t. the max of each class; Qnorm: a quantile normalization is applied. Library preprocessCore is used;
S	a named flat scores matrix with examples on rows and classes on columns.

Details

To apply the quantile normalization the preprocessCore package must be installed.

Value

the matrix of the scores flat normalized w.r.t. MaxNorm or Qnorm.

Examples

data(scores);
norm.types <- c("MaxNorm","Qnorm");
for(norm.type in norm.types){
    scores.normalization(norm.type=norm.type, S=S)
}

gecko515/HEMDAG documentation built on Oct. 18, 2019, 6:34 a.m.