7th International Conference on Communications and Networking in China

Research Article

BCE: a Privacy-preserving Common-friend Estimation Method for Distributed Online Social Networks Without Cryptography

  • @INPROCEEDINGS{10.1109/ChinaCom.2012.6417478,
        author={Yongquan Fu and Yijie Wang},
        title={BCE: a Privacy-preserving Common-friend Estimation Method for Distributed Online Social Networks Without Cryptography},
        proceedings={7th International Conference on Communications and Networking in China},
        publisher={IEEE},
        proceedings_a={CHINACOM},
        year={2012},
        month={9},
        keywords={distributed social network bloom filter privacy semi-honest model},
        doi={10.1109/ChinaCom.2012.6417478}
    }
    
  • Yongquan Fu
    Yijie Wang
    Year: 2012
    BCE: a Privacy-preserving Common-friend Estimation Method for Distributed Online Social Networks Without Cryptography
    CHINACOM
    IEEE
    DOI: 10.1109/ChinaCom.2012.6417478
Yongquan Fu1,*, Yijie Wang1
  • 1: National Key Laboratory for Parallel and Distributed Processing, College of Computer Science, National University of Defense Technology
*Contact email: yongquanf@nudt.edu.cn

Abstract

Distributed online social networks (DOSN) have emerged recently. Nevertheless, recommending friends in the distributed social networks has not been exploited fully. We propose BCE (Bloom Filter based Common-Friend Estimation), a scalable and privacy-preserving common-friend estimation scheme that estimates the set of common friends without the need of cryptography techniques. First, BCE denotes each user using the identifiers created by the Peer-to-Peer underlay that are robust against the dictionary attacks. Second, BCE uses a Bloom filter to represent a friend list for scalability. Third, BCE estimates common friends of two users using the intersection of Bloom filters computed by one of their common friends, which ensures the privacy of friend lists against unknown users. Our privacy analysis shows that BCE hides the privacy of each user with a high probability. Simulations over real-world social-network data sets confirms that BCE is both accurate and scalable.