The Feature-First Block Model
This work addresses the inference of feature impacts on network macro-structure for applications in science and engineering, representing an incremental advancement in generative models for labelled networks.
The authors tackled the problem of inferring how vertex labels affect network structure in labelled networks by introducing the feature-first block model (FFBM) and a Bayesian framework with a two-level MCMC approach, enabling efficient sampling and feature ranking without specifying explicit numbers for performance gains.
Labelled networks are an important class of data, naturally appearing in numerous applications in science and engineering. A typical inference goal is to determine how the vertex labels (or features) affect the network's structure. In this work, we introduce a new generative model, the feature-first block model (FFBM), that facilitates the use of rich queries on labelled networks. We develop a Bayesian framework and devise a two-level Markov chain Monte Carlo approach to efficiently sample from the relevant posterior distribution of the FFBM parameters. This allows us to infer if and how the observed vertex-features affect macro-structure. We apply the proposed methods to a variety of network data to extract the most important features along which the vertices are partitioned. The main advantages of the proposed approach are that the whole feature-space is used automatically and that features can be rank-ordered implicitly according to impact.