We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.SI

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Social and Information Networks

Title: Community Detection in Weighted Multilayer Networks with Ambient Noise

Abstract: We introduce a novel class of stochastic blockmodel for multilayer weighted networks that accounts for the presence of a global ambient noise governing between-block interactions. We induce a hierarchy of classifications in weighted multilayer networks by assuming that all but one cluster (block) are governed by unique local signals, while a single block behaves identically as interactions across differing blocks (ambient noise). Hierarchical variational inference is employed to jointly detect and typologize blocks as signal or noise. We call this model for multilayer weighted networks the Stochastic Block (with) Ambient Noise Model(SBANM) and develop an associated community detection algorithm. Then we apply this method to subjects in the Philadelphia Neurodevelopmental Cohort to discover communities of subjects with similar psychopathological symptoms in relation to psychosis.
Comments: 20 pages
Subjects: Social and Information Networks (cs.SI); Physics and Society (physics.soc-ph); Applications (stat.AP); Methodology (stat.ME); Machine Learning (stat.ML)
Cite as: arXiv:2103.00486 [cs.SI]
  (or arXiv:2103.00486v3 [cs.SI] for this version)

Submission history

From: Mark He [view email]
[v1] Wed, 24 Feb 2021 07:47:28 GMT (1052kb,D)
[v2] Tue, 2 Mar 2021 05:05:35 GMT (2027kb,D)
[v3] Sat, 1 May 2021 15:08:18 GMT (1628kb,D)

Link back to: arXiv, form interface, contact.