A COMPLETE PRIVACY PRESERVATION SYSTEM FOR DATA MINING USING FUNCTION APPROXIMATION
Keywords:
Privacy preservation, Radial basis function, Function approximation, Data anonymizationAbstract
Data privacy has become the primary concern in the current scenario as there are many pioneering methods for efficient mining of data. There are many algorithms to preserve privacy and handle the trade-off between privacy and utility. The ultimate goal of these algorithms is to anonymize the data without reducing the utility of them. A Privacy preserving procedure should have a minimum execution time, which is the overhead of clustering algorithms implemented using classical methods. There is also no single procedure that completely handles the trade-off and also updates itself automatically. In this work, the anonymization is implemented using Radial Basis Function [RBF] network, which provides both maximum privacy and utility with a proper tuning parameter specified between privacy and utility. The network also updates itself when the trend of data changes by controlling the maximum amount of error with a threshold value.
Downloads
References
Aggarwal. C.C, Yu.P.S (2004) : A CondensationBasedApproachtoPrivacyPreserving Data
Mining.Proceedings of the EDBT Conference, pp. 183--199.
Aggarwal, Charu C., and S. Yu Philip (2008): A general survey of privacy-preserving data
miningmodels and algorithms. Springer US, DOI : 10.1007/978-0-387-70992-5_2
Buratovic, Ines, Mario Milicevic, and KrunoslavZubrinic. (2012):Effects of data
anonymizationonthe data miningresults. MIPRO, Proceedings of the 35th International
Convention. IEEE, pp.1619 – 1623.
Byun, Ji-Won (2007): Efficient k-anonymizationusingclusteringtechniques. Advances in
Databases: Concepts, Systems and Applications. SpringerBerlin Heidelberg, pp. 188-200.
Chen, K., &Liu, L. (2009): Privacy-preservingmultipartycollaborativeminingwithgeometric data
perturbation. Parallel and DistributedSystems, IEEE Transactionson, vol.20(12), pp. 1764-1776.
Chiu, Chuang-Cheng, and Chieh-Yuan Tsai (2007): A k-anonymityclusteringmethodforeffective
data privacypreservation. Advanced Data Mining and Applications. SpringerBerlin Heidelberg,
pp. 89-99.
Dhiraj, S. S., Khan, A., Khan, W., &Challagalla, A. (2009): Privacypreservation in kmeansclusteringbyclusterrotation.
In TENCON 2009-2009 IEEE Region 10 Conference, pp. 1-7.
Evfimievski. A., Srikant. R., Agrawal. R. and Gehrke. J. (2004), “Privacypreservingmining of
association rules of InformationSystems, Vol. 29, No.4, pp. 343-364.
Fong. P. K., and Weber-Jahnke. J. H. (2012), “Privacy preserving decision tree learning using
unrealized data sets”, IEEE TransactionsonKnowledge and Data Engineering, Vol. 24, No.2,
pp.353-364.
Fung, Benjamin CM, Ke Wang, and Philip S. Yu. (2007): Anonymizingclassification data
forprivacypreservation. Knowledge and Data Engineering, IEEE Transactions on19.5, pp. 711-
, DOI: 10.1109/TKDE.2007.1015.
Ghinita, Gabriel, PanosKalnis, and Yufei Tao (2011): Anonymouspublication of
sensitivetransactional data.Knowledge and Data Engineering, IEEE Transactionson 23.2,pp.
-174,DOI: 10.1109/TKDE.2010.101.
Gionis, Aristides, and TamirTassa.(2009): k-Anonymizationwithminimalloss of
information. Knowledge and Data Engineering, IEEE Transactionson 21.2, DOI: 206-
,10.1109/TKDE.2008.129.
Guo, Kun, and Qishan Zhang (2013) : Fastclustering-basedanonymizationapproacheswith time
constraintsfor data streams.Knowledge-Based Systems46, pp. 95-108.
Han, S., Ng, W. K., Wan, L., & Lee, V. (2010). Privacy-preservinggradientdescentmethods.
Knowledge and Data Engineering, IEEE Transactions on,22(6),pp. 884-899.
Hong, D., &Mohaisen, A. (2010). AugmentedRotation-BasedTransformationforPrivacy-
Preserving Data Clustering. ETRI Journal, 32(3), pp.351-361.
Islam, M. Z., &Brankovic, L. (2011). Privacypreserving data mining: A
noiseadditionframeworkusing a novel clusteringtechnique. Knowledge-BasedSystems, 24(8), pp.
-1223.
Kisilevich, Slava, et al. "Efficient multidimensional suppressionfor k-anonymity."Knowledge and
Data Engineering, IEEE Transactionson 22.3 (2010): 334-347,DOI : 10.1109/TKDE.2009.91.
Kumar, Pradeep, KishoreIndukuriVarma, and AshishSureka (2011):
Fuzzybasedclusteringalgorithmforprivacypreserving data mining. International Journal of
Business InformationSystems 7.1,pp. 27-40.
Li, Ninghui, Tiancheng Li, and SureshVenkatasubramanian.(2010): "Closeness: A new
privacymeasurefor data publishing." Knowledge and Data Engineering, IEEE
Transactionson 22.7 ,pp. 943-956, DOI:10.1109/TKDE.2009.139.
Li, T., & Li, N. (2008): Towardsoptimal k-anonymization. Data &KnowledgeEngineering, 65(1),
pp.22-39.
Li, T., Li, N., Zhang, J., &Molloy, I. (2012): Slicing: A new approachforprivacypreserving data
publishing. Knowledge and Data Engineering, IEEE Transactionson, 24(3), pp.561-574.
Li, Yaping, et al.(2012): "Enablingmultilevel trust in privacypreserving data mining." Knowledge
and Data Engineering, IEEE Transactionson 24.9, pp. 1598-1612,DOI:10.1109/TKDE.2011.124
Lin, C. J., Chen, C. H., & Lee, C. Y. (2004): A self-adaptive quantum radial
basisfunctionnetworkforclassificationapplications. In Neural Networks, 2004. Proceedings. 2004
IEEE International JointConferenceon ,Vol. 4, pp. 3263-3268.
Lin, Jun-Lin, and Meng-ChengWei.(2009): Geneticalgorithm-basedclusteringapproachfor k –
anonymization. ExpertSystemswithApplications 36.6, pp. 9784-9792.
Liu. Q., Shen. H., and Sang, Y. (2015), “Privacy-preserving data
publishingformultiplenumericalsensitiveattributes”, TsinghuaScience and Technology, Vol. 20,
No.3, pp.246-254.
Ni, Weiwei, and Zhihong Chong.(2012) : Clustering-orientedprivacy-preserving data
publishing. Knowledge-BasedSystems , Vol.35 , pp. 264-270.
Oliveira, Stanley RM, and O. Zaiane. (2004) : Data perturbationbyrotationforprivacypreservingclustering.
TechnicalReport TR04, Vol. 17.
Rajalakshmi.V, &AnandhaMala.G. S.(2014): IsometricRelocation of Data bySequencing of Sub-
ClustersforPrivacyPreservation in Data Mining. International Journal of
Engineering&Technology), Vol.6, Issue 2.
Rebollo-Monedero, David, Jordi Forne, and Josep Domingo-Ferrer.(2010): From t-closenesslikeprivacyto
post randomizationviainformationtheory. Knowledge and Data Engineering, IEEE
Transactionson 22.11, pp.1623-1636,DOI: 10.1109/TKDE.2009.190.
Samet, S., &Miri, A. (2012). Privacy-preserving back-propagation and extreme learning machine
algorithms. Data &KnowledgeEngineering, Vol. 79, pp. 40-61.
Sun. X., Sun. L., and Wang. H. (2011), “Extended kanonymitymodelsagainstsensitiveattributedisclosure”,
ComputerCommunications, Vol.34, No.4,
pp. 526-535.
Tamersoy, A., Loukides, G., Nergiz, M. E., Saygin, Y., &Malin, B. (2012). Anonymization of
longitudinal electronic medical records. InformationTechnology in Biomedicine, IEEE
Transactionson, 16(3) , pp. 413-423.
Tao, Yufei, HekangChen, XiaokuiXiao, ShuigengZhou, and Donghui Zhang.(2009): Angel:
Enhancingtheutility of generalizationforprivacypreservingpublication.Knowledge and Data
Engineering, IEEE Transactionson 21, no. 7 ,pp. 1073-1087,DOI: 10.1109/TKDE.2009.65.
Tsiafoulis, S. G., &Zorkadis, V. C. (2010): A Neural Network
ClusteringBasedAlgorithmforPrivacyPreserving Data Mining. InComputationalIntelligence and
Security (CIS), 2010 International Conferenceon, pp. 401-405.
Vaidya, J., & Clifton, C. (2004): Privacy-preserving data mining: Why, how, and when. IEEE
Security &Privacy, 2(6), pp.19-27.
Wahlstrom. K., Roddick. J. F., Sarre. R., Estivill-Castro. V., and de Vries. D. (2009), “Legal and
technicalissues of privacypreservation in data mining”, Encyclopedia of Data Warehousing and
Mining, SecondEdition, pp. 1158-1163.
Wang, Jian, (2009): "A surveyonprivacypreserving data mining." DatabaseTechnology and
Applications, First International Workshopon. IEEE, DOI : 10.1109/DBTA.2009.147.
