Differentially Private Network Data Release via Structural Inference thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Differentially Private Network Data Release via Structural Inference

Published on Oct 07, 20142094 Views

Information networks, such as social media and email networks, often contain sensitive information. Releasing such network data could seriously jeopardize individual privacy. Therefore, we need to san

Related categories

Chapter list

Differentially Private Network Data Release via Structural Inference00:00
Idea Spotlight - 100:02
Idea Spotlight - 200:11
Idea Spotlight - 300:16
Why Privacy-aware Network Data Release ? - 100:31
Why Privacy-aware Network Data Release ? - 201:27
Problem Statement - 102:22
Problem Statement - 203:02
Problem Statement - 303:28
State-of-the-art Approaches 03:45
Our Approach: Differentially Private Network Data Release via Structural Inference04:18
Outline - 105:06
Hierarchical Random Graph05:07
Why HRG ? - 106:21
Why HRG ? - 206:48
Why HRG ? - 307:37
HGR space - 108:04
HGR space - 208:17
Outline - 208:33
What to do with HRG ? MCMC process - 108:34
What to do with HRG ? MCMC process - 208:55
What to do with HRG ? MCMC process - 309:19
Structure Inference under DP with MCMC - 109:47
Structure Inference under DP with MCMC - 210:09
Structure Inference under DP with MCMC - 310:26
Structure Inference under DP with MCMC - 410:35
Outline - 310:52
Sensitivity Analysis10:53
Outline - 411:26
Datasets11:28
MCMC Convergence Study on log ℒ - 111:38
MCMC Convergence Study on log ℒ - 211:47
Degree distribution12:56
Shortest path length distribution13:18
Overlap of top-k vertices13:24
Mean absolute error of top-k vertices13:26
Outline - 513:30
Conclusion13:31
References14:08
Thank you!14:10