getOptimalCentroids: getOptimalCentroids

In muHVT: Constructing Hierarchical Voronoi Tessellations and Overlay Heatmap for Data Analysis

getOptimalCentroids

Description

Get Optimal Centroids

Usage

getOptimalCentroids(
  x,
  iter.max,
  algorithm,
  n_cells,
  function_to_calculate_distance_metric,
  function_to_calculate_error_metric = c("mean", "max"),
  quant.err,
  distance_metric = "L1_Norm",
  quant_method = c("kmeans", "kmedoids"),
  ...
)

Arguments

x	Data Frame. A dataframe of multivariate data. Each row corresponds to an observation, and each column corresponds to a variable. Missing values are not accepted.
algorithm	String. The type of algorithm used for quantization. Available algorithms are Hartigan and Wong, "Lloyd", "Forgy", "MacQueen". (default is "Hartigan-Wong")
n_cells	Numeric. Indicating the number of nodes per hierarchy.
function_to_calculate_distance_metric	Function. The function is to find 'L1_Norm" or "L2_Norm" distances. L1_Norm is selected by default.
function_to_calculate_error_metric	Character. The error metric can be "mean" or "max". mean is selected by default
quant.err	Numeric. The quantization error for the algorithm.
distance_metric	Character. The distance metric to calculate inter point distance. It can be 'L1_Norm" or "L2_Norm". L1_Norm is selected by default.
quant_method	Character. The quant_method can be "kmeans" or "kmedoids". kmeans is selected by default
depth	Numeric. Indicating the hierarchy depth (or) the depth of the tree (1 = no hierarchy, 2 = 2 levels, etc..)

Details

The raw data is first scaled and this scaled data is supplied as input to the vector quantization algorithm. Vector quantization technique uses a parameter called quantization error. This parameter acts as a threshold and determines the number of levels in the hierarchy. It means that, if there are 'n' number of levels in the hierarchy, then all the clusters formed till this level will have quantization error equal or greater than the threshold quantization error. The user can define the number of clusters in the first level of hierarchy and then each cluster in first level is sub-divided into the same number of clusters as there are in the first level. This process continues and each group is divided into smaller clusters as long as the threshold quantization error is met. The output of this technique will be hierarchically arranged vector quantized data.

Value

values	List. A list showing observations assigned to a cluster.
maxQE	List. A list corresponding to maximum QE values for each cell.
meanQE	List. A list corresponding to mean QE values for each cell.
centers	List. A list of quantization error for all levels and nodes.
nsize	List. A list corresponding to number of observations in respective groups.

Author(s)

Shubhra Prakash <[e-mail]>, Sangeet Moy Das <[e-mail]>

muHVT documentation built on March 7, 2023, 6:38 p.m.