Privacy-Preserving Multiparty Collaborative Mining with Geometric Data Perturbation

2009 ◽  
Vol 20 (12) ◽  
pp. 1764-1776 ◽  
Author(s):  
Keke Chen ◽  
Ling Liu
Keyword(s):  
Privacy Preserving ◽  
Data Perturbation ◽  
Geometric Data

An Improved Multi-Parameters Perturbation Privacy Preserving Association Rules Mining Algorithm

2011 ◽  
Vol 130-134 ◽  
pp. 2629-2632
Author(s):  
Jie Liu ◽  
Tian Qi Li ◽  
Jian Pei Zhang
Keyword(s):  
Association Rules ◽  
Original Data ◽  
Privacy Preserving ◽  
Frequent Itemsets ◽  
Data Perturbation ◽  
First Line ◽  
Time Efficiency ◽  
Association Rules Mining ◽  
Original Algorithm ◽  
Parameters Perturbation

Multi-parameters data perturbation method is a kind of original data perturbation methods for privacy preserving association rules mining. However, the time-efficiency of restoring the frequent itemsets in multi-parameters perturbation algorithm is still not high.One method is proposed in this paper to improve the time efficiency of multi-parameters randomized perturbation algorithm according to the characteristics of the model to restore frequent itemsets. The method improves the time efficiency by getting the elements of the first line of the inversed matrix of transformation matrix. Finally, both theoretical analysis and experimental results show that the improved algorithm is more time-efficient and space-efficient than the original algorithm.

scholarly journals A Privacy-Preserving Technique for Incremental Dataset on Cloud by Synthetic Data Perturbation

2018 ◽  
Vol 7 (3.34) ◽  
pp. 331 ◽  
Author(s):  
Vigneswari D ◽  
Kumar Kumar N ◽  
Dr R.Lakshmi Tulasi
Keyword(s):  
Electronic Health Record ◽  
Synthetic Data ◽  
Privacy Preserving ◽  
Logarithmic Transformation ◽  
Health Record ◽  
Data Perturbation ◽  
Shared Data ◽  
Technology Revolution ◽  
Electronic Health ◽  
Effective Result

Cloud is an impetus technology revolution, allowing data providers to store their privatized Electronic Health Record (EHR) for further analysis and outlook with a compromised privacy, where the shared data being exposed to various adversary attacks and malware threats. There are several masking and randomization techniques that provide hints of privacy. In this paper, the Electronic Health Record (EHR) values are perturbed using logarithmic data perturbation and outsourced to the cloud.Aims.To develop a privacy-preserving technique for effective sharing ofElectronic Health Record (EHR) on a cloud by logarithmic data perturbation.Methods.Synthetic logarithmic transformation is applied to the sensitive values in a record before they are outsourced and for better privacy.Results. Synthetic Logarithmic Transformation along with the incremental anonymization produces effective result compared to classical anonymization.. 

Sign in / Sign up

Export Citation Format

Share Document