5700
eng
reportzib
0
--
2016-01-11
--
Variational Bayesian Inference and Model Selection for the Stochastic Block Model with Irrelevant Vertices
Real World networks often exhibit a significant number of vertices which are sparsely and irregularly connected to other vertices in the network. For clustering theses networks with a model based algorithm, we propose the Stochastic Block Model with Irrelevant Vertices (SBMIV) for weighted net- works. We propose an original Variational Bayesian Expectation Maximiza- tion inference algorithm for the SBMIV which is an advanced version of our Blockloading algorithm for the Stochastic Block Model. We introduce a model selection criterion for the number of clusters of the SBMIV which is based on the lower variational bound of the model likelihood. We propose a fully Bayesian inference process, based on plausible informative priors, which is independent of other algorithms for preprocessing start values for the cluster assignment of vertices. Our inference methods allow for a multi level identification of irrelevant vertices which are hard to cluster reliably ac- cording to the SBM. We demonstrate that our methods improve on the normal Stochastic Block model by applying it to to Earthquake Networks which are an example of networks with a large number of sparsely and irregularly con- nected vertices.
1438-0064
urn:nbn:de:0297-zib-57003
Christian Tobias Willenbockel
Erlinda Körnig
Christof Schütte
ZIB-Report
16-01
eng
uncontrolled
Clustering
eng
uncontrolled
Variational Bayes EM
eng
uncontrolled
Model Selection,
eng
uncontrolled
Stochastic Block Model
eng
uncontrolled
Networks
eng
uncontrolled
unsupervised classification
eng
uncontrolled
Noise
Estimation
Classification and discrimination; cluster analysis [See also 68T10]
Numerical Mathematics
no-project
https://opus4.kobv.de/opus4-zib/files/5700/ZIB-Report_16-01.pdf