Hybrid K-means

Hybrid K-means clustering using hierarchical clustering to define cluster-centers

hybrid.kmeans(x, k = 2, hmethod = "ward.D", stat = mean, ...)

Arguments

x	A data.frame or matrix with data to be clustered
k	Number of clusters
hmethod	The agglomeration method used in hclust
stat	The statistic to aggregate class centers (mean or median)
...	Additional arguments passed to kmeans

Value

returns an object of class "kmeans" which has a print and a fitted method

Details

This method uses hierarchical clustering to define the cluster-centers in the K-means clustering algorithm. This mitigates some of the know convergence issues in K-means.

Note

options for hmethod are: "ward.D", "ward.D2", "single", "complete", "average", mcquitty", "median", "centroid"

References

Singh, H., & K. Kaur (2013) New Method for Finding Initial Cluster Centroids in K-means Algorithm. International Journal of Computer Application. 74(6):27-30

Ward, J.H., (1963) Hierarchical grouping to optimize an objective function. Journal of the American Statistical Association. 58:236-24

See also

kmeans for available ... arguments and function details

hclust for details on hierarchical clustering

Author

Jeffrey S. Evans <jeffrey_evans@tnc.org>

Examples

x <- rbind(matrix(rnorm(100, sd = 0.3), ncol = 2),
           matrix(rnorm(100, mean = 1, sd = 0.3), ncol = 2))

# Compare k-means to hybrid k-means with k=4       
km <- kmeans(x, 4)       
hkm <- hybrid.kmeans(x,k=4)       

opar <- par(no.readonly=TRUE)
par(mfrow=c(1,2))
  plot(x[,1],x[,2], col=km$cluster,pch=19, main="K-means")
  plot(x[,1],x[,2], col=hkm$cluster,pch=19, main="Hybrid K-means")
par(opar)