Semantically secure classifiers for privacy preserving data mining

M. Sumana, K. S. Hareesha, Sampath Kumar

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Essential predictions are to be made by the parties distributed at multiple locations. However, in the process of building a model, perceptive data is not to be revealed. Maintaining the privacy of such data is a foremost concern. Earlier approaches developed for classification and prediction are proven not to be secure enough and the performance is affected. This chapter focuses on the secure construction of commonly used classifiers. The computations performed during model building are proved to be semantically secure. The homomorphism and probabilistic property of Paillier is used to perform secure product, mean, and variance calculations. The secure computations are performed without any intermediate data or the sensitive data at multiple sites being revealed. It is observed that the accuracy of the classifiers modeled is almost equivalent to the non-privacy preserving classifiers. Secure protocols require reduced computation time and communication cost. It is also proved that proposed privacy preserving classifiers perform significantly better than the base classifiers.

Original languageEnglish
Title of host publicationSecurity and Privacy Management, Techniques, and Protocols
PublisherIGI Global Publishing
Pages66-95
Number of pages30
ISBN (Electronic)9781522555841
ISBN (Print)1522555838, 9781522555834
DOIs
Publication statusPublished - 06-04-2018

Fingerprint

Data mining
Classifiers
Communication
Costs

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Cite this

Sumana, M., Hareesha, K. S., & Kumar, S. (2018). Semantically secure classifiers for privacy preserving data mining. In Security and Privacy Management, Techniques, and Protocols (pp. 66-95). IGI Global Publishing. https://doi.org/10.4018/978-1-5225-5583-4.ch003
Sumana, M. ; Hareesha, K. S. ; Kumar, Sampath. / Semantically secure classifiers for privacy preserving data mining. Security and Privacy Management, Techniques, and Protocols. IGI Global Publishing, 2018. pp. 66-95
@inbook{36f857e419a649db95253724f93cb256,
title = "Semantically secure classifiers for privacy preserving data mining",
abstract = "Essential predictions are to be made by the parties distributed at multiple locations. However, in the process of building a model, perceptive data is not to be revealed. Maintaining the privacy of such data is a foremost concern. Earlier approaches developed for classification and prediction are proven not to be secure enough and the performance is affected. This chapter focuses on the secure construction of commonly used classifiers. The computations performed during model building are proved to be semantically secure. The homomorphism and probabilistic property of Paillier is used to perform secure product, mean, and variance calculations. The secure computations are performed without any intermediate data or the sensitive data at multiple sites being revealed. It is observed that the accuracy of the classifiers modeled is almost equivalent to the non-privacy preserving classifiers. Secure protocols require reduced computation time and communication cost. It is also proved that proposed privacy preserving classifiers perform significantly better than the base classifiers.",
author = "M. Sumana and Hareesha, {K. S.} and Sampath Kumar",
year = "2018",
month = "4",
day = "6",
doi = "10.4018/978-1-5225-5583-4.ch003",
language = "English",
isbn = "1522555838",
pages = "66--95",
booktitle = "Security and Privacy Management, Techniques, and Protocols",
publisher = "IGI Global Publishing",
address = "United States",

}

Sumana, M, Hareesha, KS & Kumar, S 2018, Semantically secure classifiers for privacy preserving data mining. in Security and Privacy Management, Techniques, and Protocols. IGI Global Publishing, pp. 66-95. https://doi.org/10.4018/978-1-5225-5583-4.ch003

Semantically secure classifiers for privacy preserving data mining. / Sumana, M.; Hareesha, K. S.; Kumar, Sampath.

Security and Privacy Management, Techniques, and Protocols. IGI Global Publishing, 2018. p. 66-95.

Research output: Chapter in Book/Report/Conference proceedingChapter

TY - CHAP

T1 - Semantically secure classifiers for privacy preserving data mining

AU - Sumana, M.

AU - Hareesha, K. S.

AU - Kumar, Sampath

PY - 2018/4/6

Y1 - 2018/4/6

N2 - Essential predictions are to be made by the parties distributed at multiple locations. However, in the process of building a model, perceptive data is not to be revealed. Maintaining the privacy of such data is a foremost concern. Earlier approaches developed for classification and prediction are proven not to be secure enough and the performance is affected. This chapter focuses on the secure construction of commonly used classifiers. The computations performed during model building are proved to be semantically secure. The homomorphism and probabilistic property of Paillier is used to perform secure product, mean, and variance calculations. The secure computations are performed without any intermediate data or the sensitive data at multiple sites being revealed. It is observed that the accuracy of the classifiers modeled is almost equivalent to the non-privacy preserving classifiers. Secure protocols require reduced computation time and communication cost. It is also proved that proposed privacy preserving classifiers perform significantly better than the base classifiers.

AB - Essential predictions are to be made by the parties distributed at multiple locations. However, in the process of building a model, perceptive data is not to be revealed. Maintaining the privacy of such data is a foremost concern. Earlier approaches developed for classification and prediction are proven not to be secure enough and the performance is affected. This chapter focuses on the secure construction of commonly used classifiers. The computations performed during model building are proved to be semantically secure. The homomorphism and probabilistic property of Paillier is used to perform secure product, mean, and variance calculations. The secure computations are performed without any intermediate data or the sensitive data at multiple sites being revealed. It is observed that the accuracy of the classifiers modeled is almost equivalent to the non-privacy preserving classifiers. Secure protocols require reduced computation time and communication cost. It is also proved that proposed privacy preserving classifiers perform significantly better than the base classifiers.

UR - http://www.scopus.com/inward/record.url?scp=85047354898&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047354898&partnerID=8YFLogxK

U2 - 10.4018/978-1-5225-5583-4.ch003

DO - 10.4018/978-1-5225-5583-4.ch003

M3 - Chapter

AN - SCOPUS:85047354898

SN - 1522555838

SN - 9781522555834

SP - 66

EP - 95

BT - Security and Privacy Management, Techniques, and Protocols

PB - IGI Global Publishing

ER -

Sumana M, Hareesha KS, Kumar S. Semantically secure classifiers for privacy preserving data mining. In Security and Privacy Management, Techniques, and Protocols. IGI Global Publishing. 2018. p. 66-95 https://doi.org/10.4018/978-1-5225-5583-4.ch003