Tuesday, 27 April 2010

Stopping Twitter spambots & social networking spam

Interesting paper by HP researcher Miranda Mowbray, entitled The Twittering Machine, for the 6th International Conference on Web Information Systems and Technologies, 7-10 April 2010, Valencia, Spain -

"This paper is a study of the use of Twitter by automated agents, based on data sampled in July-September 2009. It discusses the dramatic rise in rapidly-tweeting automated Twitter accounts beginning in late June 2009; some surprising behaviour by automated Twitter profiles that make direct use of Twitter’s API; and techniques used for automated spamming on Twitter. Ideas are suggested for ways in which Twitter might defend against some common types of automated Twitter spam. The paper ends by outlining some general conclusions for designers of social information systems."

In terms of spambot-preventative lessons for designers of social networking sites, she says that "Forbidding automated use is not the solution… The solution is rather to create technical limits to the automated use of the system so as to allow nonautomated use to flourish… by increasing the cost (in money, time or human effort) of… particular automated behaviours…that decrease the usefulness of the system for non-automated users, without being essential for legitimate marketing that may provide revenue… Another observation is that to ensure that marketers do not make a nuisance of themselves in a social information network, it is not sufficient that marketing messages are opt-in only… Content validation may also help protect against some kinds of automated misbehaviour."

©WH. This work is licensed under a Creative Commons Attribution Non-Commercial Share-Alike England 2.0 Licence. Please attribute to WH, Tech and Law, and link to the original blog post page. Moral rights asserted.