Privacy Preserving Data Mining Essay

737 Words2 Pages

I. LITERATURE SURVEY Privacy Preserving Data Mining (PPDM) was proposed by D. Agrawal and C. C. Agrawal [1] and by Y. Lindell and B. Pinkas [5] simultaneously. To address this problem, researchers have since proposed various solutions that fall into two broad categories based on the level of privacy protection they provide. The first category of the Secure Multiparty Computation (SMC) approach provides the strongest level of privacy; it enables mutually distrustful entities to mine their collective data without revealing anything except for what can be inferred from an entity’s own input and the output of the mining operation alone by Y. Lindell and B. Pinkas in [5], J. Vaidya and C.W.Clifton in [6]. In principle, any data mining algorithm can be implemented by using generic algorithms of SMC by O.Goldreich in [7].However, these algorithms are extraordinarily expensive in practice, and impractical for real use. To avoid the high computational cost, various solutions those are more efficient than generic SMC algorithms have been proposed for specific mining tasks. Solutions to build decision trees over the horizontally partitioned data were proposed by Y. Lindell and B. Pinkas in [5]. For vertically partitioned data, algorithms have been proposed to address the association rule mining by J. Vaidya and C.W.Clifton in [6], k-means clustering by J. Vaidya and C. Clifton in[8], and frequent pattern mining problems by A.W.C. Fu, R.C.W. Wong, and K. Wang in [9]. The work of by B. Bhattacharjee, N. Abe, K. Goldman, B. Zadrozny, V.R. Chillakuru, M.del Carpio, and C. Apte in [10] uses a secure coprocessor for privacy preserving collaborative data mining and analysis. The second category of the partial information hiding approach trades pr... ... middle of paper ... ... that the encoding system by W. K. Wong, D. W. Cheung, E. Hung, B. Kao, and N. Mamoulis in [24] can be broken without using context-specific information. The success of the attacks in [25] mainly relies on the existence of unique, common, and fake items, defined by W. K. Wong, D. W. Cheung, E. Hung, B. Kao, and N. Mamoulis in [24]; our scheme does not create any such items, and the attacks by Y. Lindell and B. Pinkas in [5] are not applicable to our scheme. Tai et al. [9] assumed the attacker knows exact frequency of single items, similarly to us.

More about Privacy Preserving Data Mining Essay

Open Document