Morteza Haghir Chehreghani, Mostafa Haghir Chehreghani
UAI, New-York City, USA, June 25-29, 2016.
An important source of high clustering coefficient
in real-world networks is transitivity. However,
existing algorithms which model transitivity
suffer from at least one of the following problems:
i) they produce graphs of a specific class
like bipartite graphs, ii) they do not give an analytical
argument for the high clustering coefficient
of the model, and iii) their clustering coefficient
is still significantly lower than real-world
networks. In this paper, we propose a new model
for complex networks which is based on adding
transitivity to scale-free models. We theoretically
analyze the model and provide analytical arguments
for its different properties. In particular,
we calculate a lower bound on the clustering coefficient
of the model which is independent of
the network size, as seen in real-world networks.
More than theoretical analysis, the main properties
of the model are evaluated empirically and
it is shown that the model can precisely simulate
real-world networks from different domains with
and different specifications.
Report number: