HDclust (version 1.0.4)

HDclust-package: Clustering high dimensional data with Hidden Markov Model on Variable Blocks

Description

Clustering of high dimensional data with Hidden Markov Model on Variable Blocks (HMM-VB) fitted via Baum-Welch algorithm. Clustering is performed by the Modal Baum-Welch algorithm (MBW), which finds modes of the density function.

Author

Lin Lin, Yevhen Tupikov, Lixiang Zhang and Jia Li.Maintainer: Jia Li

Details

For a quick introduction to HDclust see the vignette vignette("HDclust").

References

Lin Lin and Jia Li, "Clustering with hidden Markov model on variable blocks," Journal of Machine Learning Research, 18(110):1-49, 2017.

See Also

hmmvbTrain, hmmvbClust

Examples

data("sim3")
set.seed(12345)
Vb <- vb(2, dim=40, bdim=c(10,30), numst=c(3,5), varorder=list(c(1:10),c(11:40)))
hmmvb <- hmmvbTrain(sim3[,1:40], VbStructure=Vb)
clust <- hmmvbClust(sim3[,1:40], model=hmmvb)
show(clust)