Dynamic Centrality for Directed Co-author Network with Context

Alisa Babskova (1) , Jan Martinovič (1) , Pavla Dráždilová (1) , Kateřina Slaninová (1)

1 VSB - Technical University of Ostrava [Ostrava]

Co-author network is a typical example of dynamic complex network, which evolves and changes over time. One of the ways how to capture and describe the dynamics of the network is determination of Stationarity for detected communities in the network. In the paper, we have proposed the modified Stationarity, which is focused only on co-authors of a given author and not on the whole community to which the author belongs. Therefore, this modified Stationarity is defined for each author in the network and is perceived as dynamic centrality. The relations in homogeneous co-author network are not only set by the number of common publications, but are given by a context to terms used by the author extracted from the article titles. This dynamic centrality calculates with the evaluation by context of directed edges in co-author network. Such modified Stationarity gives us information about stability or dynamics of the author’s neighbourhood that influences her/him, or about the stability and dynamics of the author’s neighbourhood, which the author influences in relation to context.

Mots clés: Co-author Network; Directed Network; Context; Dynamic Networks; Stationarity

Domaines: Informatique [cs]; Sciences de l'information et de la communication

Alisa Babskova, Jan Martinovič, Pavla Dráždilová, Kateřina Slaninová. Dynamic Centrality for Directed Co-author Network with Context. 13th IFIP International Conference on Computer Information Systems and Industrial Management (CISIM), Nov 2014, Ho Chi Minh City, Vietnam. pp.165-174, ⟨10.1007/978-3-662-45237-0_17⟩. ⟨hal-01405576⟩