Thursday, October 16, 2014 3:00 pm
-
3:00 pm
EDT (GMT -04:00)
Detecting Communities by Voting Model
Speaker: | Yu. Nesterov |
---|---|
Affiliation: | CORE/UCL, Belgium |
Room: | Mathematics and Computer Building (MC) 5136B |
Abstract:
In
this
talk
we
analyze
a
voting
model
based
on
random
preferences
of
participants.
Each
voter
can
choose
a
party
with
certain
probability,
which
depends
on
the
divergence
between
his
preferences
and
a
flexible
position
of
the
party.
Our
model
represents
a
rare
example
of
a
community
detection
model
with
unique
equilibrium
solution.
We
propose
an
efficient
algorithm
for
finding
the
equilibrium
state.