SLOMS: A Privacy Preserving Data Publishing Method for Multiple Sensitive Attributes Microdata

摘要

Multi-dimension bucketization is a typicalmethod to anonymize multiple sensitive attributes. However,the method leads to low data utility when microdata havemore sensitive attributes. In addition, the methods do notgeneralize quasi-identifiers, which make the anonymousdata vulnerable to suffer from linked attacks. To addressthe problems, the paper proposes a SLOMS method. Themethod vertically partitions the multiple sensitive attributesinto several tables and bucketizes each sensitive attributetable to implement l-diversity. At the same time, itgeneralizes the quasi-identifiers to implement k-anonymity.The paper also proposes a MSB-KACA algorithm toanonymize microdata with multiple sensitive attributes bySLOMS. Experiments show that SLOMS can generateanonymous tables with less suppression ratio and lessdistortion compared with generalization and MSB.