Blockmodeling is a method for clustering units in a network.

Generalized blockmodeling is an deterministic direct blockmodeling approach. Several approaches to generalized blockmodeling exist, including:

• Generalized blockmodeling of binary networks
• Generalized blockmodeling of valued networks
• Homogeneity blockmodeling

In addition to blockmodeling one-mode and two-mode networks, it can also be used for blockmodeling linked networks.

I am developing and maintaining blockmodeling R package

Here is a search result my publications relented to generalized blockmodeling from COBISS

I am developing and creating an R package kmBlock for k-means blockmodeling of one-mode and linked networks.

The package is available on R-Forge together with some other blockmodeling R packages. To install directly within R type:

• For kmBlock: install.packages("kmBlock",repos="http://R-Forge.R-project.org")
• For kmBlockTest: install.packages("kmBlockTest",repos="http://R-Forge.R-project.org")

Relevant paper:

• ŽIBERNA, Aleš. K-means-based algorithm for blockmodeling linked networks. Social Networks. [Print ed.]. May 2020, vol. 61, str. 153-169, ilustr. ISSN 0378-8733. https://www.sciencedirect.com/science/article/pii/S0378873319300176, DOI: 10.1016/j.socnet.2019.10.006.
• CUGMAS, Marjan, MALI, Franc, ŽIBERNA, Aleš. Scientific collaboration of researchers and organizations : a two-level blockmodeling approach. Scientometrics. Dec. 2020, vol. 125, no. 3, str. 2471-2489, ilustr. ISSN 0138-9130. DOI: 10.1007/s11192-020-03708-x.

For general information see Stochastic_block_model I am developing and creating an R package kmBlock for k-means blockmodeling of one-mode and linked networks.

• EUSN 2021 Workshop: Generalized blockmodeling in R using blockmodeling package