Modeling Link Recommendations as a Network Growth Mechanism and their Impact on Social Contagion

Björn Komander, Jesus Cerquides, Jeffrey Chan, Azadeh Alavi

Track: long paper (up to 10 pages)

Keywords: Link Recommendations, Social Dynamics, Graph neural networks (GNNs), Social Networks, Network Evolution

TL;DR: We model link recommendations as network growth mechanisms in a synthetic network and analyze their impact on social contagion, showing that complex contagions are highly sensitive to clustering- and homophily-based recommendations

Abstract:

Link recommendation algorithms significantly shape online social networks, in-
fluencing both their structural evolution and critical processes such as informa-
tion and behavior spread. This paper investigates how these algorithms affect
simple and complex contagion processes by modeling recommendations as addi-
tional network growth mechanisms. We introduce a synthetic network model that
integrates preferential attachment, triadic closure, and choice homophily, then ex-
tend it with various link recommenders, including heuristics and graph neural net-
works (GNNs). Our findings show that while simple contagions exhibit relatively
modest shifts under most recommenders, complex contagions are highly sensitive
to clustering- and homophily-based recommendations, thriving at moderate rec-
ommendation strengths but sharply diminishing under excessive recommendation
strength. These results underscore the nuanced interplay between network struc-
ture, recommendation strength, and contagion dynamics, highlighting the impor-
tance of incorporating social contagions into the design of link recommendation
algorithms

Submission Number: 16