HySoN: A Distributed Agent Gossip Protocol for Group Formation in Social Networks Fabrizio Messina and Giuseppe Pappalardo and Domenico Rosaci and Corrado Santoro and Giuseppe M.L. Sarn´ e Abstract On-line social networks allows people to easily interact with each other by means of social computer services. This scenario makes possible to search in a social network for affinities or new opportunities that satisfy specific requirements. However, for many users such activities often imply undesirable accesses to per- sonal sensitive data. In this scenario we propose a novel approach, called HySoN, based on an overlay network of software agents, which exploits a gossip proto- col. HySoN allows users to locally maintain sensitive user’s data, satisfying the privacy requirements preserving sensitive data. Indeed, the properties involved in the HySoN user aggregation are inferred by local data not published in the social network. Some experimental results obtained on simulated on-line social networks data show the searching of suitable nodes is very efficient due to the topology of the overlay network, which exhibits the small-world properties. 1 Introduction The rapid diffusion of general purpose social networks like Facebook 1 , MySpace 2 , Twitter 3 , and thematic networks as, for instance, Flickr 4 or Linkedin 5 , is represent- Fabrizio Messina, Giuseppe Pappalardo, Corrado Santoro Dipartimento di Matematica e Informatica, Universit` a di Catania, V.le Andrea Doria 6, 95125 Catania, Italy, e-mail: {fmessina,pappalardo,santoro}@dmi.unict.it Domenico Rosaci, Giuseppe M.L. Sarn´ e {DIIES, DICEAM}, University “Mediterranea” of Reggio Calabria, Via Graziella, Feo di Vito 89122 Reggio Calabria (RC), Italy, e-mail: {domenico.rosaci,sarne}@unirc.it 1 www.facebook.com 2 www.myspace.com 3 www.twitter.com 4 www.flickr.com 5 www.linkedin.com 1 This is the authors' version of the paper. The final publication is available at Springer, via DOI:10.1007/978-3-642-40776-5_27