MHD: A New Method towards Privacy Protecting Datasets Published
2012 ◽
Vol 214
◽
pp. 792-798
Keyword(s):
In this paper, we proposed a multi-hierarchical diversity algorithm MHD to prevent privacy disclosing in dataset. We proposed some definitions of multi-hierarchical diversity firstly. Sensitive values are partitioned into several classes. We ensured no proportion of class exceeding the threshold. We generalized some values of sensitive attribute to reduce information loss. Clustering method was used to lower data distort. Greed algorithm was used to lower time cost. We compared MHD with classic algorithms, ε-cloning and m-Invariance about Time Cost, Data Distort, Usability and Imbalance. Empirical results showed that our algorithm could protect privacy and publish datasets with high security and lower information loss
2013 ◽
Vol 457-458
◽
pp. 793-796
Keyword(s):
Keyword(s):
Keyword(s):
2012 ◽
Vol 17
(2)
◽
pp. 176-197
◽
2019 ◽
Vol 6
(6)
◽
pp. 258-263
2013 ◽
Vol 4
(3)
◽
pp. 813-820
2016 ◽
Vol 8
(2)
◽
pp. 1-11
Keyword(s):