Singular value decomposition based data distortion strategy for privacy protection

Xu, Shuting; Zhang, Jun; Han, Deping; Wang, Jie

doi:10.1007/s10115-006-0001-2

Cited by 78 publications

(52 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If there is any confidential and sensitive data, a nonnegative matrixfactorization technique can be employed to protect the original data and preserve its overall characteristics (please see details about matrix-factorization technique in Wu et al [38] and Xu el at. [39]). Airline industries are using such an information sharing mechanism in their businesses [31].…”

Section: Valuable Mechanism For Sharing Informationmentioning

confidence: 93%

Incentive information sharing in various market structures

Yan

Pei

2015

Decision Support Systems

View full text Add to dashboard Cite

Section: Valuable Mechanism For Sharing Informationmentioning

confidence: 93%

Incentive information sharing in various market structures

Yan

Pei

2015

Decision Support Systems

View full text Add to dashboard Cite

“…We used the privacy measures of the PPDM methods based on matrix factorization [9][10][11]. We assumed that the original data are denoted by and the modified data are denoted by .…”

Section: Privacy Measuresmentioning

confidence: 99%

“…Only part of the PPDM methods are not algorithm relevant, which includes two main kinds: the k-anonymity model [4][5][6] and methods based on matrix decompositions and transformation [9][10][11], both are perturbation-based methods. They overcome the shortcoming of the algorithm-relevant methods.…”

Section: Introductionmentioning

confidence: 99%

A New Algorithm-independent Method for Privacy-Preserving Classification Based on Sample Generation

Li¹,

Meng²

2015

TOCSJ

View full text Add to dashboard Cite

Abstract:With the development of data mining technologies, privacy protection is becoming a challenge for data mining applications in many fields. To solve this problem, many PPDM (privacy-preserving data mining) methods have been proposed. One important type of PPDM method is based on data perturbation. Only part of the data-perturbation-based methods is algorithm-irrelevant, which are favorable because common data mining algorithms can be used directly. This paper proposes a new algorithm-irrelevant PPDM method for classification based on sample generation. This method is a data-perturbation-based method and has three steps. First, it trains classifiers use the original data. Then, it generates new samples as the perturbed data randomly. Finally, it use the classifiers trained in the first step to predict these samples' category. The experiments show that this new method can produce usable data while protecting privacy well.

show abstract

“…To address the privacy problem various privacy preserving data mining methods in both centralized and distributed database environment have been discussed by authors in [4]. The authors in [5] proposed a Singular Value Decomposition (SVD) strategy for privacy preservation. Further they discussed some metrics to measure the difference between distorted dataset and the original dataset and the degree of privacy protection.…”

Section: Literature Surveymentioning

confidence: 99%