Int J Game Theory (2018) 47:155–173
How to identify experts in a community?
Accepted: 14 June 2017 / Published online: 21 June 2017
© Springer-Verlag GmbH Germany 2017
Abstract The group identiﬁcation literature mostly revolves around the problem of
identifying individuals in the community who belong to ethnic or religious groups.
Here we use the same model framework to identify individuals who play key role
in some sense. In particular we will focus on expert selection in social networks.
Ethnic groups and expert groups need completely different approaches and different
type of selection rules are successful for one and for the other. We argue that stabil-
ity is a key property in expert selection. The idea is that experts are more effective
in identifying each other, thus the selected individuals should support each other’s
membership. We propose a parametric algorithm based on the so called top candidate
relation. The parameter expresses how permissive we want to be in expert selection.
The two limit cases are the stable set and the top candidate core. The former contains
virtually everybody that can be considered as an expert, while the latter consists of
the elite. We establish an axiomatization to show that the algorithm is theoretically
well-founded. Furthermore we present a case study using citation data to demonstrate
its effectiveness. We compare its performance with classical centrality measures.
Keywords Group identiﬁcation · Expert selection · Stability · Citation analysis
The group identiﬁcation problem is a special votingsituationwherethesetofvotersand
the set of alternatives coincide, and the preferences are incomplete and dichotomous.
Game Theory Research Group, Centre for Economics and Regional Science, Hungarian
Academy of Sciences Budaörsi 45, Budapest 1112, Hungary
Corvinus University of Budapest, F˝ovám tér 8, Budapest 1093, Hungary