kolesnikov: Quantatization error modeling

In mattmail/clusterAnalysis: Several tools for determining the number of clusters in a dataset

Description

This method is based on the within cluster variance W. The idea is to estimate a parameter a and to find the minimum of this function : W(k)k^a.

Usage

kolesnikov(X, maxK, clusterAlg = myKmean)

Arguments

X	data matrix or data frame of size n x d, n observations and d features
maxK	maximum number of clusters to evaluate
clusterAlg	clustering algorithm. Its output must be a list containing parameters "cluster" and "center". For more details, check the formatting of function myKmean.

Value

list having 3 attributes:

kopt

optimal number of clusters

PCF

vector of score

a

value of the exponent a

References

Kolesnikov, A., Trichina, E., and Kauranne, T. (2015). Estimating the number of clusters in a numerical data set via quantizationerror modeling.Pattern Recognition, 48:941-952.

mattmail/clusterAnalysis documentation built on Nov. 4, 2019, 6:18 p.m.