2016 Fiscal Year Final Research Report
A Positive Study on Evidence Based Anonymization
| Project/Area Number |
26380269
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Economic statistics
|
|---|
| Research Institution | Kanazawa University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Keywords | 個票開示リスク / 離散分布論 |
|---|
| Outline of Final Research Achievements |
The information protection of an individual is legally defined as unidentifiability. However, this definition is technically vague, and it often leads to the overanonymization of microdata. To mitigate this social loss, we need a clear technical definition of identifiability. Our study aims to clarify this unidentifiability as an object of statistical estimation, and has constructed the detailed theory of this idea. Also our study notes that the exting Anonymized Data have the information of unidentifiability of the data, and thus has measured the possibility of identification on these data of Housing and Land Survey of 2003.
|
| Free Research Field |
経済統計学
|
