| Abstract | There are currently, thousands of online social networks (OSN)available, each of which hosts millions of users. Users make new
friendship links, join groups on diverse topics, share opinions and
thus help in building a big knowledge repository. In this paper we
study the problem of defining proximity measure between groups
(communities) of OSN. Understanding the proximity among the
groups in OSN can reveal new insights into the structure of the
network. We use this proximity measure in a novel way with other
structural features of online groups to predict the evolving state of
the network. One of the important task in the sphere of OSN is
helping users in selecting groups of their interest, by making rec-
ommendations. We utilize group proximity measure to propose a
new framework for recommending groups to users of OSN. Finally
we investigate the effectiveness of our methods in three real OSN:
Flickr, Live Journal and You Tube. Our methods give promising
results.
|
