MKMeans {MKMeans}

Modern K-Means clustering.

Description

It's a Modern K-Means clustering algorithm allowing data of any number of dimensions, any initial center, and any number of clusters to expect.

Usage

MKMeans(data, K, initial, iteration, tol, type)

Arguments

data

Numeric. An observation matrix with each row being an oberservation.

K

Integer. The number of clusters expected.

initial

Numeric. Either the selected initial center matrix with each row being an observation, or 1 for the first K rows of the data matrix being the intial center.

iteration

Integer. The number of the most iterations wanted for the clustering process.

tol

Numeric. The minimum acceptable percentage of stable observations to stop the clustering process, basically greater than 0.5 to guarantee the value of the results.

type

Integer. The type of distance between observations. 1 for Euclidean distance. 2 for Manhattan distance. 3 for maximum deviation among dimensions.

Value

An object of class MKMean.

Author(s)

Yi Ya

References

Yarong Yang(Yi Ya) and Jacob Zhang.(2022) MKMeans: A Modern K-Means Clustering Algorithm. submitted to Journal of American Statistical Association

Examples

x<-rnorm(20,0,1)
y<-rnorm(20,1,1)
data.test<-cbind(x,y)
Res<-MKMeans(data.test,3,1,iteration=1000,tol=.95,type=1)
Ress<-Res
names(Ress@Classes[[1]])<-rep("red",length(Res@Classes[[1]]))
names(Ress@Classes[[2]])<-rep("blue",length(Res@Classes[[2]]))
names(Ress@Classes[[3]])<-rep("green",length(Res@Classes[[3]]))
Cols<-names(sort(c(Ress@Classes[[1]],Ress@Classes[[2]],Ress@Classes[[3]])))
plot(x,y,type="p",col=Cols,lwd=2)
points(Res@Centers,pch=15,col=c("red","blue","green"))