Article

Privacy homomorphisms for social networks with private relationships

Rovira i Virgili University, Department of Computer Engineering and Mathematics, UNESCO Chair in Data Privacy, Av. Països Catalans 26, Tarragona 43007, Catalonia, Spain
Computer Networks (Impact Factor: 1.28). 10/2008; 52(15):3007-3016. DOI: 10.1016/j.comnet.2008.06.017
Source: DBLP

ABSTRACT Enabling private relationships in social networks is an important issue recently raised in the literature. We describe in this paper a new protocol which offers private relationships allowing resource access through indirect relationships without requiring a mediating trusted third party (although an optimistic trusted third party is used which only acts in case of conflict). Thanks to homomorphic encryption, our scheme prevents the resource owner from learning the relationships and trust levels between the users who collaborate in the resource access. In this way, the number of users who might refuse collaboration due to privacy concerns is minimized. This results in increased resource availability, as the chances that certain nodes become isolated at a given period of time are reduced. Empirical evidence is provided about the proposed protocol being scalable and deployable in practical social networks.

Download full-text

Full-text

Available from: Alexandre Viejo, Aug 23, 2015
0 Followers
 · 
119 Views
    • "Significant work has been done exploring cryptographic approaches to enhance the content sharing privacy on OSNs. Work in [11] presents a publickey protocol which achieves relationship protection without the presence of a central node so enabling private relationships using certificates or verifiable credentials. A number of the proposed access control models leverage users' attributes . "
    IET Intelligent Transport Systems 04/2015; 9(3):293-305. DOI:10.1049/iet-its.2013.0202 · 0.95 Impact Factor
  • Source
    • "For instance, [22] proposed a decentralized solution for path discovery based on public key cryptography, but the user triggering the path discovery can learn all properties of each relationship between two nodes in the path. To overcome this drawback in [23] authors enhanced the protocol with hormomophic encryption so users are able only to access aggregated value of relationship properties. In [24], two indirect nodes can compute the depth of all the paths between them without referring to other nodes. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The technological advancements in Internet speeds, increased computing power and smart phones have pushed the rise of new digital methods supporting mobile person-to-person (P2P) payments. Despite the growing interest in these new methods, we be-lieve that, to fully enable this increasing rise of digital wallets, there is the need for tools helping a person in judging the risk of a money transfer. For this purpose, this paper aims at exploiting social net-work connections. This is achieved by making pay-ers/payees able to state their trust preferences with respect to the potential payees/payers. Trust pref-erence evaluation requires to find social connections between a payer and a payee across, possible, differ-ent social network realms. We therefore propose a light cryptography protocol, specifically targeted to mobile P2P payments, that besides providing good performance, ensures user information privacy.
  • Source
    • "For instance, [3] proposed a decentralized solution for path discovery based on public key cryptography, but the user triggering the path discovery can learn all properties of each relationship between two nodes in the path. To overcome this drawback in [4] authors enhanced the protocol with hormomophic encryption so users are able only to access aggregated value of relationship properties. In [13], two indirect nodes can compute the depth of all the paths between them without referring to other nodes. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The technological advancements in Internet speeds, increased computing power and smart phones have pushed the rise of new digital methods supporting mobile person-to-person (P2P) payments. Despite the growing interest in these new methods, we believe that, to fully enable this increasing rise of digital wallets, there is the need for tools helping a person in judging the risk of a money transfer. For this purpose, this paper aims at exploiting social network connections. This is achieved by making payers/payees able to state their trust preferences with respect to the potential payees/payers. Trust preference evaluation requires to find social connections between a payer and a payee across, possible, different social network realms. We therefore propose a light cryptography protocol, specifically targeted to mobile P2P payments, that besides providing good performance, ensures user information privacy.
    Social Computing (SocialCom), 2013 International Conference on; 01/2013
Show more