***** To join INSNA, visit http://www.insna.org ***** Here are two papers which discuss detecting sock puppets on Twitter (but not necessarily identify their real operators) are. Might give you some ideas, though. Detecting and Tracking Political Abuse in Social Media Ratkiewicz, J., Conover, M.D., Meiss, M., Gonçalves, B., Flammini, A., and Menczer, F. 2011. In Proceedings of the Fifth International AAAI Conference on Weblogs and Social Media. This papers describes the Truthy project which was inspired by the Twitterbomb concept coined by Mustafaraj and Metaxas . Truthy is a system to detect astroturf political campaigns either to simulate widespread support for a candidate or to spread disinformation. The paper describes in reasonable detail the system, providing a number of real case examples and performance analysis. http://truthy.indiana.edu/site_media/pdfs/ratkiewicz_icwsm2011_truthy.pdf Barash, V. Kelly, J. Salience Vs. Commitment:Dynamics of Political Hashtags in Russian Twitter, 6th Annual Network Science Workshop, 22-24 April 2012, West Point, NY http://www.westpoint.edu/nsc/SiteAssets/SitePages/Workshops/Barash%20Political%20Hashtags%20in%20Russian%20Twitter.pdf Brian Ulicny VIStology, Inc. On Tue, Oct 2, 2012 at 2:15 PM, Devin Gaffney <[log in to unmask]> wrote: > ***** To join INSNA, visit http://www.insna.org ***** > Hey all, > > A friend recently brought up an interesting question - there's a user in > their local interest "community" on Twitter that's being particularly > disruptive - was there any way that I could possibly figure out the identity > of the person? Obviously, it's a much more complex question than just > ferreting it out, and I don't know if it's even ethical to ferret out the > pseudonymous account's owner, but it is an interesting question that I would > love to hear methodology on. Basically, the situation is as such: given a > pseudonymous account (operated by a person who is likely a member of the > community, and likely has their own account within that community), is there > any way to use a network approach to yield likely candidate "real" accounts > that may also be operated by that person? Implications are pretty clear here > for use in disambiguating dissidents and such, so it's likely not a totally > safe road to travel down too far, but it is an interesting question that I > would love to hear thoughts on. My gut was saying that there could be some > way to measure observed/expected triads for the user and its ties going > outward maybe 2 degrees - if you could establish a short list of people that > they were likely to be friends with, but for some reason are not, this could > be possibly useful. > > Anyways, any thoughts on how to approach that question would be useful, from > what sort of theoretical assumptions could/should be made, ethical issues, > to possible methodologies. > > Devin > > _____________________________________________________________________ SOCNET > is a service of INSNA, the professional association for social network > researchers (http://www.insna.org). To unsubscribe, send an email message to > [log in to unmask] containing the line UNSUBSCRIBE SOCNET in the body of > the message. _____________________________________________________________________ SOCNET is a service of INSNA, the professional association for social network researchers (http://www.insna.org). To unsubscribe, send an email message to [log in to unmask] containing the line UNSUBSCRIBE SOCNET in the body of the message.