期刊名称:International Journal of Computer Science and Network Security
印刷版ISSN:1738-7906
出版年度:2008
卷号:8
期号:1
页码:115-121
出版社:International Journal of Computer Science and Network Security
摘要:Data privacy is the most acclaimed problem when publishing individual data. It ensures individual data publishing without disclosing sensitive data. The much popular approach, is K-Anonymity, where data is transformed to equivalence classes, each class having a set of K- records that are indistinguishable from each other. But several authors have pointed out numerous problems with K-anonymity and have proposed techniques to counter them or avoid them. l-diversity and t-closeness are such techniques to name a few. Our study has shown that all these techniques increase computational effort to practically infeasible levels, though they increase privacy. A few techniques account for too much of information loss, while achieving privacy. In this paper, we propose a novel, holistic approach for achieving maximum privacy with no information loss and minimum overheads (as only the necessary tuples are transformed). We address the data privacy problem using fuzzy set approach, a total paradigm shift and a new perspective of looking at privacy problem in data publishing. Our practically feasible method in addition, allows personalized privacy preservation, and is useful for both numerical and categorical attributes.
关键词:Privacy preserving, data privacy, fuzzy information, anonymity