Precisely Answering Multi-Dimensional Range Queries without Privacy Breach
This paper studies the privacy breaches caused by multi-dimensional range (MDR) sum queries in online analytical processing (OLAP) systems. We show that existing inference control methods are generally infeasible for controlling MDR queries. We then consider restricting users to even MDR queries (th...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Institutional Knowledge at Singapore Management University
2003
|
| Subjects: | |
| Online Access: | https://ink.library.smu.edu.sg/sis_research/556 http://dx.doi.org/10.1007/b13237 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Singapore Management University |
| Language: | English |
| Summary: | This paper studies the privacy breaches caused by multi-dimensional range (MDR) sum queries in online analytical processing (OLAP) systems. We show that existing inference control methods are generally infeasible for controlling MDR queries. We then consider restricting users to even MDR queries (that is, the MDR queries involving even numbers of data values). We show that the collection of such even MDR queries is safe if and only if a special set of sum-two queries (that is, queries involving exactly two values) is safe. On the basis of this result, we give an efficient method to decide the safety of even MDR queries. Besides safe even MDR queries we show that any odd MDR query is unsafe. Moreover, any such odd MDR query is different from the union of some even MDR queries by only one tuple. We also extend those results to the safe subsets of unsafe even MDR queries. |
|---|