Hi everyone,
Today, Liangzhe talked about two papers:
1. "On Integrating Network and Community Discovery" (WSDM 2015): this
paper talked about how to integrate community detection with network
discovery. The problem is, given Gs(Ns,As,Qs), a target node set Nt
(subset of Ns), an ability to query any currently observe node for their
adjacent links at cost ci, we want to cluster Nt into the set of k most
tightly linked communities within a total budget B.
There may be some useful evaluations for us such as purity and entropy
used in this paper. And we can also think about the problem like
community detection+inferring network structure.
2. "Global Diffusion via Cascading Invitations: Structure, Growth and
Homophily" (WWW 2015): this papers investigated diffusion patterns
through invitation from existing members in Linkedin.
There are some interesting new patterns of diffusion in Linkedin, e.g.,
the cascade size in Linkedin goes linearly w.r.t. the time which is
different from Twitter and Facebook.
Next Thursday, if Ravi is not available for the talk, Elaheh will give a
presentation.
Thank you.
Regards,
Yao