clustDDist-internal: Internal Functions in the clustDDist package

In clustDDist: Clustering Discrete Distributions

Description

Internal functions of the package.

Usage

compute.clusters(x, centers, err.measure, penalty = 1e+06)
compute.D(i, j, leader, err.measure, size, sum, harmonic, penalty, ...)
compute.hleader(i, j, leader, err.measure, size, sum, harmonic, penalty, ...)
compute.leaders(x, k, curr_clusters, err.measure, penalty = 1e+06)
d1.cleader(cluster_units, ...)
d1.D(size1, size2, error, ...)
d1.hleader(leader1, leader2,size1, size2, ...)
d1(unit, centers, penalty = 1e+06)
diss.matrix(leader, err.measure, size, sum, harmonic, penalty, ...)
harmonic.sum(x, penalty = 1e+06)
hclustering(x, err.measure, penalty = 1e+06, nleaders, fromLeaders)
leaders.algorithm(x, centers, err.measure = "d4", stabil = 1e-06, penalty = 1e+06, echo = FALSE)
sort.matrix.rows(m)
update.diss(index_min, diss, leader, err.measure, size, sum, harmonic, penalty, ...)
update.leader(mini, maxi, leader, err.measure, size, sum, harmonic, penalty)
update.matrix(mini, maxi, mat)
update.vector(mini, maxi, vec)

Arguments

x	Dataframe or matrix of units (in rows). Rows can represent probabilities or frequencies.
centers	matrix of current leaders
err.measure	String for error function that will be used to calculate the error between current leader and a unit. Possibilities are from d1 to d7 (see reference).
penalty	Parameter to specify how should division by zero be treated.
i	i-th cluster
j	j-th cluster
leader	current leader matrix
size	vector - current size of clusters
sum	vector - current sum of cluster units
harmonic	vector - current harmonic sum of cluster units
k	number of clusters
curr_clusters	vector of cluster numbers for each unit
cluster_units	matrix of units in current cluster
size1	size of first cluster
size2	size of second cluster
leader1	vector - leader of first cluster
leader2	vector - leader of second cluster
error	error among two clusters (computed acc. to error function)
unit	vector - a unit
nleaders	How many leaders should be saved (NULL or between 1 and n (number of units)). Note: leaders cannot be always computed after the whole procedure is done.
fromLeaders	NULL or list that one gets from function from.Leaders in order to take size and units from leaders method into consideration.
stabil	Stability parameter - used in order to achieve convergence if too many clustering iterations.
m	numeric matrix to be sorted
index_min	vector of 2 indices - two cluster numbers to be merged
diss	current dissimilarity matrix
mini	lowest cluster number to be merged
maxi	highest cluster number to be merged
mat	numeric matrix
vec	numeric vector
echo	if TRUE function prints error of each iteration
...	other possible parameters

Details

The functions are written just for error function d1. Similarly there are functions for other error functions (d2 to d7). Internal functions are used to break the code into a more easily readable functions.

clustDDist documentation built on May 2, 2019, 6:47 p.m.