| Project/Area Number |
23K24851
|
|---|
| Project/Area Number (Other) |
22H03595 (2022-2023)
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Multi-year Fund (2024)
Single-year Grants (2022-2023) |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 60080:Database-related
|
|---|
| Research Institution | Institute of Science Tokyo (2024)
Hokkaido University (2022-2023) |
|---|
Principal Investigator |
曹 洋 東京工業大学, 情報理工学院, 准教授 (60836344)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
吉川 正俊 大阪成蹊大学, データサイエンス学部, 教授 (30182736)
小西 葉子 関西学院大学, 総合政策学部, 専任講師 (00876708)
鄭 舒元 大阪大学, 大学院情報科学研究科, 特任助教(常勤) (30994694)
|
|---|
| Project Period (FY) |
2022-04-01 – 2025-03-31
|
| Project Status |
Granted (Fiscal Year 2024)
|
| Budget Amount *help |
¥17,160,000 (Direct Cost: ¥13,200,000、Indirect Cost: ¥3,960,000)
Fiscal Year 2024: ¥6,500,000 (Direct Cost: ¥5,000,000、Indirect Cost: ¥1,500,000)
Fiscal Year 2023: ¥6,500,000 (Direct Cost: ¥5,000,000、Indirect Cost: ¥1,500,000)
Fiscal Year 2022: ¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
|
|---|
| Keywords | differential privacy / 差分プライバシ / 説明可能性 / プライバシー保護 / Differential Privacy / Explainability / 11111 |
|---|
| Outline of Research at the Start |
Our project aims at providing a crucial component to the existing DP systems: a principled framework for explaining, choosing, a nd negotiating privacy parameters in differentially private analysis. We call such a framework Explainable DP, which is a set of approaches, guidelines, and toolkits.
|
| Outline of Annual Research Achievements |
We conduct research on privacy budgeting in the context of differentially private data analysis. Our focus areas include differentially private trajectory event stream publishing, spatiotemporal data releasing, differentially private streaming data release, and locally private streaming data release with shuffling and subsampling.
Preliminary experimental results indicate that the main factor affecting the trade-off between privacy risk and utility metrics is the specific scenario in which the analysis is conducted. A key takeaway from our research is that achieving an optimal privacy budget is not universally possible without considering the specific differentially private algorithms and data distributions involved.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
Our findings have been published in reputable venues such as IEEE MDM, IEEE ICDE workshops, IEEE CSF, and the IEICE journal.
|
| Strategy for Future Research Activity |
We will now move on to our third topic of privacy budgeting: how to reconcile conflicts when different stakeholders have varying requirements for privacy parameters. Our basic idea is to use data market mechanisms.
|
