gmsClust: A general implementation of Modha-Spangler clustering for...

In kamila: Methods for Clustering Mixed-Type Data

Description

Modha-Spangler clustering estimates the optimal weighting for continuous vs categorical variables using a brute-force search strategy.

Usage

gmsClust(
  conData,
  catData,
  nclust,
  searchDensity = 10,
  clustFun = wkmeans,
  conDist = squaredEuc,
  catDist = squaredEuc,
  ...
)

Arguments

conData	A data frame of continuous variables.
catData	A data frame of categorical variables; the allowable variable types depend on the specific clustering function used.
nclust	An integer specifying the number of clusters.
searchDensity	An integer determining the number of distinct cluster weightings evaluated in the brute-force search.
clustFun	The clustering function to be applied.
conDist	The continuous distance function used to construct the objective function.
catDist	The categorical distance function used to construct the objective function.
...	Arguments to be passed to the clustFun.

Details

Modha-Spangler clustering uses a brute-force search strategy to estimate the optimal weighting for continuous vs categorical variables. This implementation admits an arbitrary clustering function and arbitrary objective functions for continuous and categorical variables.

The input parameter clustFun must be a function accepting inputs (conData, catData, conWeight, nclust, ...) and returning a list containing (at least) the elements cluster, conCenters, and catCenters. The list element "cluster" contains cluster memberships denoted by the integers 1:nclust. The list elements "conCenters" and "catCenters" must be data frames whose rows denote cluster centroids. The function clustFun must allow nclust = 1, in which case $centers returns a data frame with a single row. Input parameters conDist and catDist are functions that must each take two data frame rows as input and return a scalar distance measure.

Value

A list containing the following results objects:

results	A results object corresponding to the base clustering algorithm
objFun	A numeric vector of length searchDensity containing the values of the objective function for each weight used
Qcon	A numeric vector of length searchDensity containing the values of the continuous component of the objective function
Qcon	A numeric vector of length searchDensity containing the values of the categorical component of the objective function
bestInd	The index of the most successful run
weights	A numeric vector of length searchDensity containing the continuous weights used

References

Foss A, Markatou M; kamila: Clustering Mixed-Type Data in R and Hadoop. Journal of Statistical Software, 83(13). 2018. doi: 10.18637/jss.v083.i13

Modha DS, Spangler WS; Feature Weighting in k-Means Clustering. Machine Learning, 52(3). 2003. doi: 10.1023/a:1024016609528

Examples

## Not run: 
# Generate toy data set with poor quality categorical variables and good
# quality continuous variables.
set.seed(1)
dat <- genMixedData(200, nConVar=2, nCatVar=2, nCatLevels=4, nConWithErr=2,
  nCatWithErr=2, popProportions=c(.5,.5), conErrLev=0.3, catErrLev=0.8)
catDf <- dummyCodeFactorDf(data.frame(apply(dat$catVars, 2, factor), stringsAsFactors = TRUE))
conDf <- data.frame(scale(dat$conVars), stringsAsFactors = TRUE)

msRes <- gmsClust(conDf, catDf, nclust=2)

table(msRes$results$cluster, dat$trueID)

## End(Not run)

Example output

   
     1  2
  1 89 23
  2 13 75

kamila documentation built on March 13, 2020, 9:08 a.m.