https://github.com/aaamini/mbisbm
Matched BIpartite SBM (with node covariates/metadata)
Science Score: 10.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
○codemeta.json file
-
○.zenodo.json file
-
○DOI references
-
✓Academic publication links
Links to: arxiv.org -
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (4.6%) to scientific vocabulary
Last synced: 10 months ago
·
JSON representation
Repository
Matched BIpartite SBM (with node covariates/metadata)
Basic Info
- Host: GitHub
- Owner: aaamini
- Language: MATLAB
- Default Branch: master
- Size: 1.65 MB
Statistics
- Stars: 2
- Watchers: 3
- Forks: 1
- Open Issues: 0
- Releases: 0
Created over 7 years ago
· Last pushed over 7 years ago
https://github.com/aaamini/mbisbm/blob/master/
# mbisbm `mbisbm` stands for Matched BIpartite Stochastic Block Model. The code here implements algorithms in the paper entitled [Matched bipartite block model with covariates](https://arxiv.org/abs/1703.04943). The main function is `common/fit_mbiSBM.m` which implements the latest version of the algorithm. This is the defulat version and should be used if you have a good initialization. See `example_wiki_cities.m` and `example_wiki_topart.m` for how to use this function. `common/biSpecClust.m` implements the matched bipartite clustering described in the paper. This can be used as the initialization for the mbisbm or as a standalone algorithm. There is also `common/fit_mbiSBM_v2.m` which implements v.2 of the algorithm. This is used for example in creating some of the early figures in the paper. You can run `simulations/compare_nmi_sims.m` to see it in action. V.2 is more robust to the initial labeling. Use this version if you have unreliable/bad initial labels.
Owner
- Name: Arash A. Amini
- Login: aaamini
- Kind: user
- Company: @ucla
- Website: http://www.stat.ucla.edu/~arashamini
- Repositories: 4
- Profile: https://github.com/aaamini