This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
class:privacy2022sgrad [2022/03/03 11:54] mhshin [Overview] |
class:privacy2022sgrad [2025/10/13 12:45] (current) |
||
|---|---|---|---|
| Line 4: | Line 4: | ||
| * **Provided by**: Dept. of Computer Engineering, Myongji University | * **Provided by**: Dept. of Computer Engineering, Myongji University | ||
| * **Lead by**: Minho Shin (mhshin@mju.ac.kr, Rm5742) | * **Lead by**: Minho Shin (mhshin@mju.ac.kr, Rm5742) | ||
| - | * **Period**: Fall semester, 2017 | + | * **Period**: Spring semester, 2022 |
| * **Location**: 5701 at 5th Engineering Building | * **Location**: 5701 at 5th Engineering Building | ||
| * **Time**: Wednesdays, 1pm to 4pm | * **Time**: Wednesdays, 1pm to 4pm | ||
| Line 16: | Line 16: | ||
| ====== Agenda ====== | ====== Agenda ====== | ||
| - | ^ Date ^ Name ^ Topic ^ Slides ^ Data ^ Method ^ Evaluation ^ Contribution | ||
| - | | 2017.*.* | Minho | Topics in Data Privacy |{{ :class:gpriv2017f:lecture-privacy.pptx |}} \\ {{ :class:gpriv2017f:differentialprivacy2.pptx |}} |{{ :class:gpriv2017f:data_new.zip |}} | | | | ||
| - | ====== Term Project ====== | ||
| - | * Goal: Analyze the privacy levels of data using ARX + alpha | + | ====== Term Project ====== |
| - | * Steps | + | |
| - | - Choose your data (already done) | + | |
| - | - 권유진: adult | + | |
| - | - Zhong: atus | + | |
| - | - 장민해: cup | + | |
| - | - 이수정: fars | + | |
| - | - 노건: ihis | + | |
| - | - Download your data and VHG | + | |
| - | - Open the data in ARX | + | |
| - | - Choose quasi-identifiers | + | |
| - | - Choose one sensitive attribute | + | |
| - | - Import VHGs | + | |
| - | - Check if VHG is valid. If not, fix it. | + | |
| - | - Generalization steps should be in detail | + | |
| - | - Run de-identification algorithms | + | |
| - | - run k-anonymity with k=3, 5, 10 | + | |
| - | - choose suppression limit as you wish | + | |
| - | - run distinctive l-diversity with l=3, 5, 10 | + | |
| - | - run entropy l-diversity with l=3, 5, 10 | + | |
| - | - run t-closeness with your choice of t | + | |
| - | - For each de-identified data do | + | |
| - | - Compute Min/Max/Avg EQ group sizes | + | |
| - | - Compute Min/Max/Avg # of different sensitive values | + | |
| - | - Compute Min/Max/Avg entropy | + | |
| - | - Compute GenInfoLoss | + | |
| - | - Compute the privacy levels | + | |
| - | - Compute identity-disclosure level | + | |
| - | * {{:class:gpriv2017f:screen_shot_2017-12-06_at_3.18.25_am.png?150|}} | + | |
| - | - Compute attribute-disclosure level | + | |
| - | * {{:class:gpriv2017f:screen_shot_2017-12-06_at_3.22.43_am.png?200|}} | + | |
| - | - Compute inference-disclosure level | + | |
| - | * {{:class:gpriv2017f:screen_shot_2017-12-06_at_3.23.49_am.png?200|}} | + | |
| - | - Analyze the results of privacy levels | + | |
| - | - Discuss the results | + | |