A New Efficient Privacy-Preserving Scalar Product Protocol

Recently, privacy issues have become important in data analysis, especially when data is horizontally partitioned over several parties. In data mining, the data is typically represented as attribute-vectors and, for many applications, the scalar (dot) product is one of the fundamental operations that is repeatedly used. In privacy-preserving data mining, data is distributed across several parties. The efficiency of secure scalar products is important, not only because they can cause overhead in communication cost, but dot product operations also serve as one of the basic building blocks for many other secure protocols. Although several solutions exist in the relevant literature for this problem, the need for more efficient and more practical solutions still remains. In this paper, we present a very efficient and very practical secure scalar product protocol. We compare it to the most common scalar product protocols. We not only show that our protocol is much more efficient than the existing ones, we also provide experimental results by using a real life dataset.

[1]  Jaideep Vaidya,et al.  Privacy preserving association rule mining in vertically partitioned data , 2002, KDD.

[2]  Wenliang Du,et al.  Privacy-preserving cooperative statistical analysis , 2001, Seventeenth Annual Computer Security Applications Conference.

[3]  Vipin Kumar,et al.  Privacy Preserving Nearest Neighbor Search , 2006, Sixth IEEE International Conference on Data Mining - Workshops (ICDMW'06).

[4]  Silvio Micali,et al.  How to play ANY mental game , 1987, STOC.

[5]  Jesus Mena,et al.  Investigative Data Mining for Security and Criminal Detection , 2002 .

[6]  Ljiljana Brankovic,et al.  Data Swapping: Balancing Privacy against Precision in Mining for Logic Rules , 1999, DaWaK.

[7]  E. Reingold,et al.  Combinatorial Algorithms: Theory and Practice , 1977 .

[8]  Vladimir Estivill-Castro,et al.  Private representative-based clustering for vertically partitioned data , 2004, Proceedings of the Fifth Mexican International Conference in Computer Science, 2004. ENC 2004..

[9]  Artak Amirbekyan,et al.  Privacy-preserving regression algorithms , 2007 .

[10]  Chris Clifton,et al.  Privacy-preserving outlier detection , 2004, Fourth IEEE International Conference on Data Mining (ICDM'04).

[11]  Wenliang Du,et al.  Building decision tree classifier on private data , 2002 .

[12]  Mikhail J. Atallah,et al.  A secure protocol for computing dot-products in clustered and distributed environments , 2002, Proceedings International Conference on Parallel Processing.

[13]  Yunghsiang Sam Han,et al.  Privacy-Preserving Multivariate Statistical Analysis: Linear Regression and Classification , 2004, SDM.

[14]  Artak Amirbekyan,et al.  The Privacy of k-NN Retrieval for Horizontal Partitioned Data -- New Methods and Applications , 2007, ADC.

[15]  Chris Clifton,et al.  Privately Computing a Distributed k-nn Classifier , 2004, PKDD.

[16]  Chris Clifton,et al.  Privacy-Preserving Data Mining , 2006, Encyclopedia of Database Systems.

[17]  Pascal Paillier,et al.  Public-Key Cryptosystems Based on Composite Degree Residuosity Classes , 1999, EUROCRYPT.

[18]  Oded Goldreich,et al.  Foundations of Cryptography: General Cryptographic Protocols , 2004 .

[19]  Andrew Chi-Chih Yao,et al.  Protocols for secure computations , 1982, FOCS 1982.

[20]  Artak Amirbekyan,et al.  Privacy Preserving DBSCAN for Vertically Partitioned Data , 2006, ISI.

[21]  Hans-Peter Kriegel,et al.  LOF: identifying density-based local outliers , 2000, SIGMOD 2000.

[22]  Gu Si-yang,et al.  Privacy preserving association rule mining in vertically partitioned data , 2006 .

[23]  Tatsuaki Okamoto,et al.  A New Public-Key Cryptosystem as Secure as Factoring , 1998, EUROCRYPT.

[24]  Chris Clifton,et al.  Privacy-preserving k-means clustering over vertically partitioned data , 2003, KDD '03.

[25]  Jacques Stern,et al.  A new public key cryptosystem based on higher residues , 1998, CCS '98.