| 研究課題/領域番号 |
19K20269
|
|---|
| 研究機関 | 京都大学 |
|---|
研究代表者 |
曹 洋 京都大学, 情報学研究科, 特定助教 (60836344)
|
|---|
| 研究期間 (年度) |
2019-04-01 – 2022-03-31
|
| キーワード | Differential Privacy / 差分プライバシー / location privacy / 位置情報プライバシ |
|---|
| 研究実績の概要 |
We identified several insufficiencies of the state-of-the-art DP-based location privacy models. We assessed the privacy risks of the existing location privacy protection mechanisms (LPPMs) when the adversaries know road networks. It turns out that the adversaries could infer the users' location precisely.
To prevent this, we proposed countermeasures in several aspects.
First, we studied how to represent users' privacy preferences flexibly and how to formalize the secrets about the user's spatiotemporal activities. Third, we design novel mechanisms to protect users' spatiotemporal events.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
Our studies on spatiotemporal privacy have been accepted in top conferences and journal including IEEE ICDE 2019, VLDB 2019 and IEEE TKDE.
|
| 今後の研究の推進方策 |
Motivated by COVID-19, we will study how to apply our privacy techniques in the real-world for specific purpose like epidemic surveillance.
On the other hand, we will study the theoretical properties of the proposed privacy model and a generalized version of it in order to increase the applicability.
|
