Abstract | ||
---|---|---|
Anonymization is a method used in privacy-preserving data publishing. Previous studies show that anonymization based on the request of a data recipient, the priority of attributes, helps to maintain data utility. However, it is difficult for recipients to generate requests because they can not know which attribute important without data analysis. To address this issue, we propose a framework for performing custom-made anonymization by data analysis program provided by recipient. This enables the recipient to generate a request after creating a program and performing an indirect analysis of an original dataset by the program. Moreover, we describe an inference attack model for this framework and propose a secure method for restraining such an attack.
|
Year | DOI | Venue |
---|---|---|
2019 | 10.1145/3292006.3302380 | CODASPY |
Field | DocType | ISBN |
Computer science,Computer security,Inference attack,Data publishing | Conference | 978-1-4503-6099-9 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Wakana Maeda | 1 | 0 | 0.68 |
Yuji Yamaoka | 2 | 9 | 4.63 |