Community detection for dynamic networks.

Share:

Description

Dynamic network clustering/community detection using a latent space approach. Using temporal edge data, network actors are embedded onto a hypersphere and grouped based on direction.

Details

Package: dnc
Type: Package
Version: 1.0
Date: 2016-05-05
License: GPL (>= 2)

This package can perform community detection on dynamic (temporal) network data observed at discrete time points. Communities are assumed fixed, but community membership may change. The main function is dnc(...) which can perform variational Bayes estimation or alternatively implement a Gibbs sampler. A dnc object is the output, for which there exists the following generic commands: simulate(), plot(), print(), and BIC(). Ignorable (MAR, MCAR) missing edge data can be incorporated into the Gibbs sampler.

Author(s)

Daniel K. Sewell

Maintainer: Daniel K. Sewell <daniel-sewell@uiowa.edu>

References

Sewell, D. K., and Chen, Y. (2016). Latent Space Approaches to Community Detection in Dynamic Networks. Bayesian Analysis. doi: 10.1214/16-BA1000. http://projecteuclid.org/euclid.ba/1461603847

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
## Not run: 
VB5 = dnc(Y,M=5,p=3,"VB")
Gibbs5 = dnc(Y,M=5,p=3,"Gibbs")
print(VB5)
print(Gibbs5)
BIC(VB5)
BIC(Gibbs5)
plot(VB5)
plot(Gibbs5,aggregated=FALSE,plotRGL=FALSE)

## End(Not run)