An Effective and Computationally Efficient Approach for Anonymizing Large-Scale Physical Activity Data: Multi-Level Clustering-Based Anonymization
An Effective and Computationally Efficient Approach for Anonymizing Large-Scale Physical Activity Data: Multi-Level Clustering-Based Anonymization
复制标题
一种有效且计算高效的大规模体力活动数据匿名化方法：基于多级聚类的匿名化
DOI：
--
复制
发表时间：
2020
影响因子：
0.8
通讯作者：
A. G. Koru
中科院分区：
文献类型：
--
作者：
Pooja Parameshwarappa;Zhiyuan Chen;A. G. Koru
Publishingphysicalactivitydatacanfacilitatereproduciblehealth-careresearchinseveralareassuch aspopulationhealthmanagement，behavioralhealthresearch，andmanagementofchronichealth问题。However，publishingsuchdataalsobringshighprivacyrisksrelatedtore-identification whichmakesanonymizationnecessary。Oneofthechallengesinanonymizingphysicalactivitydata collectedperiodicallyisitssequentialnature。Theexistinganonymizationtechniquesworksufficiently forcross-sectionaldatabuthavehighcomputationalcostswhenapplieddirectlytosequentialdata。Thisarticlepresentsaneffectiveanonymizationapproach，multi-levelclustering-basedanonymization toanonymizephysicalactivitydata。Comparedwiththeconventionalmethods，theproposedapproach improvestimecomplexitybyreducingtheclusteringtimedrastically。Whiledoingso，itpreserves theutilityasmuchastheconventionalapproaches。
Publishingphysicalactivitydatacanfacilitatereproduciblehealth-careresearchinseveralareassuch aspopulationhealthmanagement,behavioralhealthresearch,andmanagementofchronichealth problems.However,publishingsuchdataalsobringshighprivacyrisksrelatedtore-identification whichmakesanonymizationnecessary.Oneofthechallengesinanonymizingphysicalactivitydata collectedperiodicallyisitssequentialnature.Theexistinganonymizationtechniquesworksufficiently forcross-sectionaldatabuthavehighcomputationalcostswhenapplieddirectlytosequentialdata. Thisarticlepresentsaneffectiveanonymizationapproach,multi-levelclustering-basedanonymization toanonymizephysicalactivitydata.Comparedwiththeconventionalmethods,theproposedapproach improvestimecomplexitybyreducingtheclusteringtimedrastically.Whiledoingso,itpreserves theutilityasmuchastheconventionalapproaches.
影响因子：
5
作者：
Matthews, Charles E.;Chen, Kong Y.;Troiano, Richard P.
通讯作者：
Troiano, Richard P.