English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In network inference applications, it is often desirable to detect communitystructure, namely to cluster vertices into groups, or blocks, according to somemeasure of similarity. Beyond mere adjacency matrices, many real networks alsoinvolve vertex covariates that carry key information about underlying blockstructure in graphs. To assess the effects of such covariates on blockrecovery, we present a comparative analysis of two model-based spectralalgorithms for clustering vertices in stochastic blockmodel graphs with vertexcovariates. The first algorithm uses only the adjacency matrix, and directlyestimates the block assignments. The second algorithm incorporates both theadjacency matrix and the vertex covariates into the estimation of blockassignments, and moreover quantifies the explicit impact of the vertexcovariates on the resulting estimate of the block assignments. We employChernoff information to analytically compare the algorithms' performance andderive the information-theoretic Chernoff ratio for certain models of interest.Analytic results and simulations suggest that the second algorithm is oftenpreferred: we can often better estimate the induced block assignments by firstestimating the effect of vertex covariates. In addition, real data examplesalso indicate that the second algorithm has the advantages of revealingunderlying block structure and taking observed vertex heterogeneity intoaccount in real applications. Our findings emphasize the importance ofdistinguishing between observed and unobserved factors that can affect blockstructure in graphs.

On spectral algorithms for community detection in stochastic blockmodel graphs with vertex covariates