This repository is to archive a summer research project. The project was supervised by Prof. Arash Amini at UCLA in 2015. We studied the stochastic block model with covariates (SBMC). The SBMC is a variant of the stochastic block model. It is motivated by the fact that in many network datasets, both connections between nodes and attributes of individual nodes are observed. The SBMC models the two parts and can be used to improve community detection performance.
This repository includes:
- The implementation of an alogorithm. This algorithm adapts a variational expectation–maximization (EM) approach to fit the SBMC.
- Several simulation studies that compare the algorithm to other community detection methods.
- Two data applications of the algorithm.