Skip to main content

Table 3 Models combining block models and topic models

From: A review of stochastic block models and extensions for graph clustering

Article

Type of graph

Textual

Inference

Clustering

Number of groups/topics

Longitudinal

Zhou et al. (2006)

Valued

directed

Edges

MCMC

Soft

Fixed (6 groups; 20 topics)

No

McCallum et al. (2007)

Valued

directed

Edges

MCMC

Soft

Criterion (Perplexity)

No

Pathak et al. (2008)

Valued

directed

Edges

MCMC

Soft

Fixed (8 groups; 25 topics)

No

Liu et al. (2009)

Binary

directed

Nodes

Variational

Soft

Criterion (Perplexity)

No

Chang and Blei (2010)

Binary

undirected

Nodes

Variational

Soft

Fixed (5,10,15,20,25 topics)

No

Ho et al. (2012)

Binary

directed

Nodes

MCMC

Soft

Dirichlet process

No

Sachan et al. (2012)

Valued

directed

Edges

MCMC

Soft; Soft

Criterion (Perplexity)

No

Bouveyron et al. (2016)

Valued

directed

Edges

Variational

Soft; Hard

Criterion (BIC-like)

No

Corneli et al. (2018)

Valued

directed

Edges

Variational

Soft; Hard

Criterion (ICL)

Yes

  1. Under the column “clustering”, the first approach refers to the topics. If there is a second approach, it refers to the groups