clust5 {kmed}R Documentation

5-clustered data set

Description

A dataset containing two variables of 800 objects and their class memberships generated by the clusterGeneration package.

Usage

clust5

Format

A data frame with 800 rows and 3 variables:

x1

X1.

x2

X2.

class

Class membership.

Source

Data is generated via the genRandomClust function in the clusterGeneration package. The code to generate this data set is

set.seed(2016)

randclust <- clusterGeneration::genRandomClust(5, sepVal = 0.2, numNonNoisy = 2, numReplicate = 1, clustszind = 3, clustSizes = as.numeric(table(sample(1:5, 800, replace = TRUE))), outputDatFlag=FALSE, outputLogFlag=FALSE, outputEmpirical=FALSE, outputInfo=FALSE)

clust5 <- as.data.frame(randclust$datList$test_1)

clust5$class <- randclust$memList$test_1

References

Qiu, W., and H. Joe. 2015. ClusterGeneration: Random Cluster Generation (with Specified Degree of Separation).

Qiu, W., and H. Joe. 2006a. Generation of Random Clusters with Specified Degree of Separation. Journal of Classification 23 pp. 315-34.

Qiu, W., and H. Joe. 2006b. Separation Index and Partial Membership for Clustering. Computational Statistics and Data Analysis 50 pp. 585-603.


[Package kmed version 0.4.2 Index]