fpca.cluster: Clustering the estimators along the path.

In FusedPCA: Community Detection via Fused Principal Component Analysis

Description

To use k-means to cluster the estimators along the path and get the community labels. fpca.cluster is a wrap up of functions fpca.nonscore.cluster and fpca.score.cluster. The latter two are the clustering functions for the situations in which score = FALSE and score = TRUE, respectively.

Usage

fpca.cluster(obj, K = 2, score = F)
fpca.nonscore.cluster(obj, K = 2)
fpca.score.cluster(obj, K = 2)

Arguments

obj	in function fpca.cluster, it is an object generated by fpca.start, i.e., generated by fpca.nonscore or fpca.score, if score = FALSE or score = TRUE, respectively. It is a list. In functions fpca.nonscore.cluster and fpca.score.cluster, it is an input matrix, which is a FPCA or FPCA-RoE object, of dimension number of non-isolated nodes x number of effective estimators. It is generated by fpca.nonscore and fpca.score.
K	input integer – the pre-specified number of communities, with the default value 2.
score	indicator argument – whether to apply the score associated clustering method or not, with the default value FALSE.

Value

an array of community labels list, of dimension number of non-isolated nodes x number of effective estimators. Each entry has value from 1 to K, as an index of the community label. Notice, the community labels are usually permutation-invariant.

Warning

if the input object obj is a FPCA object, the supposed value for score should be F. If users set score = T, the function will stop with warning 'This object is designed for 'score = F''. If the input object obj is a FPCA-RoE object, the supposed value for score should be T. If users set score = F, the function will still execute, but with warning 'This object is designed for 'score = T”.

Author(s)

Yang Feng, Richard J. Samworth and Yi Yu

References

Yang Feng, Richard J. Samworth and Yi Yu, Community Detection via Fused Principal Component Analysis, manuscript. Holland, P.W., Laskey, K.B. and Leinhardt, S., 1983. Stochastic block models: first steps. Social Networks 5, 109-137. Jin, J., 2012. Fast community detection by score. Karrer, B. and Newman, M.E.J., 2011. Stochastic blockmodels and community structure in networks. Physical Review E 83, 016107.

See Also

fpca.nonscore.cluster, fpca.score.cluster, fpca.start, fpca.nonscore, fpca.score.

Examples

### please see the examples in fpca

Example output

Loading required package: genlasso
Loading required package: MASS
Loading required package: Matrix
Loading required package: igraph

Attaching package: 'igraph'

The following objects are masked from 'package:stats':

    decompose, spectrum

The following object is masked from 'package:base':

    union

FusedPCA documentation built on May 29, 2017, 9:19 p.m.