2nd International ICST Conference on Scalable Information Systems

Research Article

Privacy-Preserving Statistical Quantitative Rules Mining

Download623 downloads
  • @INPROCEEDINGS{10.4108/infoscale.2007.209,
        author={Weiwei Jing and Liusheng Huang and Yifei Yao and Weijiang Xu},
        title={Privacy-Preserving Statistical Quantitative Rules Mining},
        proceedings={2nd International ICST Conference on Scalable Information Systems},
        proceedings_a={INFOSCALE},
        year={2010},
        month={5},
        keywords={Privacy-Preserving Data Mining Statistical Quantitative rules},
        doi={10.4108/infoscale.2007.209}
    }
    
  • Weiwei Jing
    Liusheng Huang
    Yifei Yao
    Weijiang Xu
    Year: 2010
    Privacy-Preserving Statistical Quantitative Rules Mining
    INFOSCALE
    ICST
    DOI: 10.4108/infoscale.2007.209
Weiwei Jing1,*, Liusheng Huang1,*, Yifei Yao1,*, Weijiang Xu2,*
  • 1: Depart.of Comp. Sci. & Tech., USTC NHPCC 416, East Campus USTC, Hefei, 230026, PRC Tel: 86-551-3602445
  • 2: Depart.of Comp. Sci. & Tech., USTC NHPCC 416, East Campus USTC, Hefei, PRC Tel: 86-551-3602445
*Contact email: wwjing@mail.ustc.edu.cn, lshuang@ustc.edu.cn, yaoyifei@mail.ustc.edu.cn, wjxu@mail.ustc.edu.cn

Abstract

This paper considers the problem of mining Statistical Quantitative rules (SQ rules) without revealing the private information of parties who compute jointly and share distributed data. Based on several basic tools for Privacy-Preserving Data Mining (PPDM), including secure sum, secure mean and secure frequent itemsets, this paper presents two algorithms to accomplish privacy-preserving SQ rules mining over horizontally partitioned data. One is to securely compute confidence intervals for testing the significance of rules; the other is to securely discover SQ rules.