Data mining techniques are rising trends to aid organizations to analyze, find un-obvious patterns and details to benefit
from the customer or user data. But this is classified as proprietary information disclosure and mining misuse. To avoid this, we
introduce the concept of privacy preserving data mining (PPDM). The fundamental notions of the existing privacy preserving data
mining methods, their merits, and shortcomings are presented. We discus five techniques namely Anonymization based PPDM,
Perturbation based PPDM, Randomized response based PPDM, Condensation based PPDM and Cryptography based PPDM.
Published In:IJCSN Journal Volume 6, Issue 5
Date of Publication : October2017
Pages : 547-550
Figures :01
Tables : --
Bhargav Sundararajan : SRM University, Chennai, Tamil Nadu 603203, India
Deepthi Peri : SRM University, Chennai, Tamil Nadu 603203, India
Nita Radhakrishnan : SRM University, Chennai, Tamil Nadu 603203, India
Mehul Awasthi : SRM University, Chennai, Tamil Nadu 603203, India
privacy, data mining, anonymization, perturbation
The primary objective of PPDM is promoting algorithm to
conceal sensitive data or over privacy. These sensitive
data do not get revealed to unapproved parties or invader.
In data mining there exists a trade of between utility and
privacy of data. When we accomplish one it inevitably
leads to the detrimental impact on the other. Many PPDM
techniques in existence are reviewed in the paper.
Ultimately, it is concluded with the fact that there is no
single PPDM technique in existence that outshines every
other technique with relation to each possible criteria such
as use of data, performance, difficulty, compatibility with
procedures for data mining, and so on. A particular
algorithm may function better when compared to another,
on a specific criterion. Various algorithms may be found
to function better than one another on given criterion.
Researchers are doing extensive research in ensuring that
the sensitive data of a person is not revealed as well as not
compromising the utility of data so that the data can be
useful for many purposes.
[1] Ann Cavoukian, Information and Privacy Commissioner,
Ontario, “Data Mining Staking a Claim on Your
Privacy”, 1997.
[2] The Economist. “The End of Privacy”, May 1st, 1999.
pp: 15.
[3] R. Agrawal and R. Srikant. “Privacy Preserving Data
Mining”, ACM SIGMOD Conference on Management of
Data, pp: 439-450, 2000.
[4] D. Agrawal and C. Aggarwal, “On the Design and
Quantification of Privacy Preserving Data Mining
Algorithms”, PODS 2001. pp: 247-255.
[5] W. Du and Z. Zhan, “Using Randomized Response
Techniques for Privacy Preserving Data Mining”,
SIGKDD 2003. pp. 505-510.
[6] Elisa, B., N.F. Igor and P.P. Loredana. “A Framework
for
Evaluating Privacy Preserving Data Mining
Algorithms”,
Published by Data Mining Knowledge Discovery, 2005,
pp.121- 154.
[7] Sweeney L, "Achieving k-Anonymity privacy protection
using generalization and suppression" International
journal of Uncertainty, Fuzziness and Knowledge based
systems, 10(5), 571- 588, 2002.
[8] Evfimievski A., "Randomization in Privacy-Preserving
Data Mining", ACM SIGKDD Explorations, 4, 2003.
[9] Aggarwal C, Philip S Yu, "A condensation approach to
privacy preserving data mining", EDBT, 183-199, 2004.
[10] Benny Pinkas,"Cryptographic Techniques for Privacy
preserving data mining", SIGKDD Explorations, Vol. 4,
Issue 2, 12-19, 2002.