It seems like cliques like this should be pretty easy to statistically discover.
Does anyone have a dataset I could try this on?
Does anyone have a dataset I could try this on?
Representative example: SybilLimit: A Near-Optimal Social Network Defense against Sybil Attacks Haifeng Yu, Phillip B. Gibbons, Michael Kaminsky,Feng Xiao http://www.comp.nus.edu.sg/~yuhf/yuh-sybillimit.pdf
A more recent paper looks at a data set from Digg to discover vote manipulation; the authors are interested in further applications and would probably appreciate your contact.
Sybil Resilient Content Rating D.M. Tran, B. Min, J. Li, L. Subramanian http://www.cs.nyu.edu/~lakshmi/nsdi09.pdf