normalize.altMethod: Alternative method for normalizing matrices

In avilesd/productConfig: Tools to analyze behavioral patterns from Product Configurators

Description

This function is used as the first step in weight.differenceToIdeal and weight.entropy for normalizing a matrix.

Usage

normalize.altMethod(aMatrix, attr, cost_ids = NULL)

Arguments

aMatrix	a numeric (decision) matrix - if matrix has only one row, then all values will be normalized to 1 even if the initial value is 0.
attr	attributes IDs, vector of integer numbers corresponding to the attributes you desire to use; attr are assumed to be 1-indexed. Even if not all attributes are given the function normalizes all columns, by default as benefit type unless shown otherwise with cost_ids.
cost_ids	argument used to convert selected cost attributes into benefit attributes. Integer vector.

Details

This functions uses an alternative approach to the one used in gainLoss in regard to cost_ids. Instead of multiplying all column values with (-1) so that lower values become the higher values, it normalizes relative to the lowest value, as given by [1] and [2]. Main difference in the result is that this normalizing function does not return negative values.

Value

a normalized matrix

References

[1]Ma, J., Fan, Z. P., & Huang, L. H. (1999). A subjective and objective integrated approach to determine attribute weights. European journal of operational research, 112(2), 397-404.

[2] Fan, Z. P. (1996). Complicated multiple attribute decision making: theory and applications (Doctoral dissertation, Ph. D. Dissertation, North-eastern University, Shenyang, PRC).

Examples

# Runnable
normalize.altMethod(matrix(1:16, 4, 4))
normalize.altMethod(matrix(1:16, 4, 4), attr=1:4, cost_ids = c(2,4))

avilesd/productConfig documentation built on May 11, 2019, 4:08 p.m.