Clustering-based k-anonymity

Xianmang He, Hua Hui Chen, Yefang Chen, Yihong Dong, Peng Wang, Zhenhua Huang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

16 Scopus citations

Abstract

Privacy is one of major concerns when data containing sensitive information needs to be released for ad hoc analysis, which has attracted wide research interest on privacy-preserving data publishing in the past few years. One approach of strategy to anonymize data is generalization. In a typical generalization approach, tuples in a table was first divided into many QI (quasi-identifier)-groups such that the size of each QI-group is no less than k. Clustering is to partition the tuples into many clusters such that the points within a cluster are more similar to each other than points in different clusters. The two methods share a common feature: distribute the tuples into many small groups. Motivated by this observation, we propose a clustering-based k-anonymity algorithm, which achieves k-anonymity through clustering. Extensive experiments on real data sets are also conducted, showing that the utility has been improved by our approach.

Original languageEnglish
Title of host publicationAdvances in Knowledge Discovery and Data Mining - 16th Pacific-Asia Conference, PAKDD 2012, Proceedings
Pages405-417
Number of pages13
EditionPART 1
DOIs
StatePublished - 2012
Externally publishedYes
Event16th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining, PAKDD 2012 - Kuala Lumpur, Malaysia
Duration: 29 May 20121 Jun 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume7301 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference16th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining, PAKDD 2012
Country/TerritoryMalaysia
CityKuala Lumpur
Period29/05/121/06/12

Keywords

  • algorithm
  • privacy preservation
  • proximity privacy

Fingerprint

Dive into the research topics of 'Clustering-based k-anonymity'. Together they form a unique fingerprint.

Cite this