Interestingness measure on privacy preserved data with horizontal partitioning

Association rule mining is a process of finding the frequent item sets based on the interestingness measure. The major challenge exists when performing the association of the data where privacy preservation is emphasized. The actual transaction data provides the evident to calculate the parameters for defining the association rules. In this paper, a solution is proposed to find one such parameter i.e. support count for item sets on the non transparent data, in other words the transaction data is not disclosed. The privacy preservation is ensured by transferring the x-anonymous records for every transaction record. All the anonymous set of actual transaction record perceives high generalized values. The clients process the anonymous set of every transaction record to arrive at high abstract values and these generalized values are used for support calculation. More the number of anonymous records, more the privacy of data is amplified. In experimental results it is shown that privacy is ensured with more number of formatted transactions.

[1]  Zoran Obradovic,et al.  A privacy-preserving framework for distributed clinical decision support , 2011, 2011 IEEE 1st International Conference on Computational Advances in Bio and Medical Sciences (ICCABS).

[2]  Panos Kalnis,et al.  Anonymous Publication of Sensitive Transactional Data , 2011, IEEE Transactions on Knowledge and Data Engineering.

[3]  Yang Tang,et al.  A Privacy Preserving Algorithm for Mining Distributed Association Rules , 2011, 2011 International Conference on Computer and Management (CAMAN).

[4]  Li Wan,et al.  Privacy-Preserving Gradient-Descent Methods , 2010, IEEE Transactions on Knowledge and Data Engineering.

[5]  Vitaly Shmatikov,et al.  How To Break Anonymity of the Netflix Prize Dataset , 2006, ArXiv.

[6]  Hong Shen,et al.  Privacy Preserving C4.5 Algorithm Over Horizontally Partitioned Data , 2006, 2006 Fifth International Conference on Grid and Cooperative Computing (GCC'06).

[7]  Ramakrishnan Srikant,et al.  Privacy-preserving data mining , 2000, SIGMOD '00.