0.25
0.5
0.75
1.25
1.5
1.75
2
Privacy and Background Knowledge
Published on Feb 25, 20073926 Views
The digitization of our daily lives has led to an explosion in the collection of data by governments, corporations, and individuals. Protection of confidentiality of this data is of utmost importance.
Chapter list
Privacy and Background Knowledge00:00
An Abundance of Data02:03
Driving Factors: A LARGE Hardware Revolution02:37
Driving Factors: A small Hardware Revolution03:12
Other Driving Factors03:39
picture04:14
picture04:44
picture05:18
Pulsars05:22
Pulsar Surveys06:12
Project Requirements06:22
Driving Factors: Analysis Capabilities06:44
And Even the Popular Press Caught On07:24
Concerns About Privacy07:44
The Setup08:27
Model I: Untrusted Data Collector08:44
Minimal Information Sharing09:15
Model II: Trusted Data Collector09:56
Disclosure Limitations10:54
Types of Disclosure11:30
Types of Disclosure11:56
Types of Disclosure12:27
Talk Outline12:51
Privacy Preserving Associations13:48
Problem Introduction14:00
Example15:01
The Itemset Lattice16:04
Frequent Itemsets16:27
Breath First Search: 1-Itemsets16:50
Breath First Search: 1-Itemsets17:01
Breath First Search: 2-Itemsets17:12
Breath First Search: 3-Itemsets17:30
Breadth First Search: Remarks17:40
Depth First Search (2)17:53
Depth First Search (3)17:55
Depth First Search (4)18:07
Talk Outline18:16
Our Model18:28
Our Model (Contd.)18:56
Our Model (Contd.)19:04
Privacy Preserving Associations19:07
Minimal Information Sharing19:29
Our Model (Contd.)19:39
Our Model (Contd.)19:47
Our Model: Another View19:51
The Problem20:24
The Randomized Response Model20:45
Another View: Two Questions21:16
Analysis21:40
Analysis (Contd.)21:55
Interval Privacy22:25
Interval Privacy: Quantifying Privacy22:46
Talk Outline23:32
Background Knowledge23:45
Example: {a, b, c}24:21
Example: {a, b, c}24:47
Example: {a, b, c}24:50
Example: {a, b, c}25:05
Example: {a, b, c}25:35
Privacy Breaches26:09
Simple Privacy Breaches26:50
Privacy Breaches: Goals27:46
α-to-β Privacy Breach28:23
α-to-β Privacy Breach28:54
α-to-β Privacy Breach29:06
α-to-β Privacy Breach29:28
α-to-β Privacy Breach29:39
Amplification Condition30:23
Amplification Condition30:35
Amplification Condition30:53
Amplification Condition31:15
Amplification Condition31:27
The Bound on α-to-β Breaches31:59
The Bound on α-to-β Breaches32:36
Amplification: Summary32:47
Talk Outline33:50
Definition of select-a-size35:42
Definition of select-a-size35:56
Definition of select-a-size36:05
Definition of select-a-size36:25
Support Recovery36:45
Support Recovery36:56
Support Recovery37:20
Support Recovery37:26
Support Recovery37:32
Support Recovery37:36
The Unbiased Estimators37:39
Apriori [AS94]37:48
The Modified Apriori38:13
Talk Outline38:27
Talk Outline38:39
Trusted Data Collector39:12
Disclosure Limitations39:25
Sample Microdata39:44
Removing SSN …40:48
Linkage Attacks41:06
Linkage Attacks (Contd.)41:42
Quasi-Identifiers and Sensitive Attributes42:32
K-Anonymity [Sweeney02]42:55
K-Anonymity43:39
K-Anonymity Through Generalization43:46
Example Microdata43:59
4-Anonymous Microdata44:08
K-Anonymity Algorithms44:39
Talk Outline44:46
Example Microdata44:55
4-Anonymous Microdata44:58
Homogeneity Attack45:02
Background Knowledge Attack45:45
Data Publishing Desiderata46:58
Privacy Definition (1)47:32
Privacy Definition (2)47:52
Bayes-Optimal Privacy– Drawbacks48:22
Towards A Practical Definition (1)48:27
Towards A Practical Definition (2)48:31
Ensuring Diversity48:34
3-Diverse Microdata49:01
L-Diversity Revisited49:28
L-Diversity: Summary49:49
What I Talked About49:55
What I Talked About (Only Useful Stuff)50:34
Open Problems51:01
Modeling Belief52:32
Thanks52:42
But Of Course We Have More Confidence Than Scott Adams …53:01
Questions?53:09