TY - JOUR
T1 - JANUS
T2 - A hypothesis-driven Bayesian approach for understanding edge formation in attributed multigraphs
AU - Espín-Noboa, Lisette
AU - Lemmerich, Florian
AU - Strohmaier, Markus
AU - Singer, Philipp
N1 - Publisher Copyright:
© 2017, The Author(s).
PY - 2017/12/1
Y1 - 2017/12/1
N2 - Understanding edge formation represents a key question in network analysis. Various approaches have been postulated across disciplines ranging from network growth models to statistical (regression) methods. In this work, we extend this existing arsenal of methods with JANUS, a hypothesis-driven Bayesian approach that allows to intuitively compare hypotheses about edge formation in multigraphs. We model the multiplicity of edges using a simple categorical model and propose to express hypotheses as priors encoding our belief about parameters. Using Bayesian model comparison techniques, we compare the relative plausibility of hypotheses which might be motivated by previous theories about edge formation based on popularity or similarity. We demonstrate the utility of our approach on synthetic and empirical data. JANUS is relevant for researchers interested in studying mechanisms explaining edge formation in networks from both empirical and methodological perspectives.
AB - Understanding edge formation represents a key question in network analysis. Various approaches have been postulated across disciplines ranging from network growth models to statistical (regression) methods. In this work, we extend this existing arsenal of methods with JANUS, a hypothesis-driven Bayesian approach that allows to intuitively compare hypotheses about edge formation in multigraphs. We model the multiplicity of edges using a simple categorical model and propose to express hypotheses as priors encoding our belief about parameters. Using Bayesian model comparison techniques, we compare the relative plausibility of hypotheses which might be motivated by previous theories about edge formation based on popularity or similarity. We demonstrate the utility of our approach on synthetic and empirical data. JANUS is relevant for researchers interested in studying mechanisms explaining edge formation in networks from both empirical and methodological perspectives.
KW - Attributed multigraphs
KW - Bayesian inference
KW - Edge formation
KW - HypTrails
KW - Multiplex
UR - http://www.scopus.com/inward/record.url?scp=85055788232&partnerID=8YFLogxK
U2 - 10.1007/s41109-017-0036-1
DO - 10.1007/s41109-017-0036-1
M3 - Article
AN - SCOPUS:85055788232
SN - 2364-8228
VL - 2
JO - Applied Network Science
JF - Applied Network Science
IS - 1
M1 - 16
ER -