System and method for database privacy protection

Patent metadata

Field	Value
Publication number	US-9355258-B2
Application number	US-201214345818-A
Country	US
Kind code	B2
Filing date	Sep 25, 2012
Priority date	Sep 28, 2011
Publication date	May 31, 2016
Grant date	May 31, 2016

Abstract

The invention relates to a system and a method for privacy preservation of sensitive attributes stored in a database. The invention reduces the complexity and enhances privacy preservation of the database by determining the distribution of sensitive data based on Kurtosis measurement. The invention further determines and compares the optimal value of k-sensitive attributes in k-anonymity data sanitization model with the optimal value of l sensitive attributes in l diversity data sanitization model using adversary information gain. The invention reduces the complexity of the method for preserving privacy by applying k anonymity only, when the distribution of the sensitive data is leptokurtic and optimal value of k is greater than the optimal value of l.

First claim

We claim:

1. A database privacy protection method comprising:
determining, via one or more processors, a distribution pattern of one or more database attributes by applying Kurtosis measurement of data corresponding to the attributes to ascertain whether the distribution pattern is leptokurtic;
determining, via the one or more processors, an adversary information gain for a k-anonymity data sanitization model and adversary information gain for a k-anonymity l-diversity data sanitization model, wherein the adversary information gain is the difference between entropy of S and a conditional entropy H(S|Q), and wherein S corresponds to a set of the attributes;
comparing, via the one or more processors, the adversary information gain of the k-anonymity data sanitization model with the adversary information gain of the k-anonymity l-diversity data sanitization model repeatedly until the adversary information gain of the k-anonymity data sanitization model equals the adversary information gain of the k-anonymity l-diversity data sanitization model;
determining, via the one or more processors, an optimal value of l for performing l-diversity based data sanitization on database records related to the attributes and an optimal value of k for performing k-anonymity based data sanitization on the attributes; and
performing, via the one or more processors, privacy preservation of the attributes by only k-anonymity data sanitization model when k is greater than l and the distribution pattern is leptokurtic.

2. The method as claimed in claim 1 , wherein the method further comprises the step of determining the optimal value of l for performing l-diversity based data sanitization and k for performing k-anonymity based data sanitization using a Normalized Certainty Penalty.

3. The method as claimed in claim 1 , wherein the method further comprises the step of determining a privacy disclosure probability that decreases with increase in value of k/l.

4. The method as claimed in claim 1 , wherein the method further comprises the step of evaluating a reduction in complexity for performing data privacy preservation by using the k-anonymity data sanitization model as compared to using the k-anonymity l-diversity data sanitization model.

5. A database privacy protection system comprising:
one or more hardware processors; and
one or more memory units storing instructions executable by the one or more hardware processors to perform operations comprising:
determining a distribution pattern of one or more database attributes by applying Kurtosis measurement of data corresponding to the attributes to ascertain whether the distribution pattern is leptokurtic;
determining an adversary information gain for a k-anonymity data sanitization model and adversary information gain for a k-anonymity l-diversity data sanitization model, wherein the adversary information gain is the difference between entropy of S and a conditional entropy H(S|Q), and wherein S corresponds to a set of the attributes;
comparing the adversary information gain of the k-anonymity data sanitization model with the adversary information gain of the k-anonymity l-diversity data sanitization model repeatedly until the adversary information gain of the k-anonymity data sanitization model equals the adversary information gain of the k-anonymity l-diversity data sanitization model;

determining an optimal value of l for performing l-diversity based data sanitization on database records related to the attributes and an optimal value of k for performing k-anonymity based data sanitization on the attributes; and
performing privacy preservation of the attributes by only k-anonymity data sanitization model when k is greater than l and the distribution pattern is leptokurtic.

6. The system as claimed in claim 5 , the one or more memory units storing instructions executable by the one or more hardware processors to perform operations further comprising:
calculating a privacy disclosure probability that decreases with increase in value of k/l.

7. The system as claimed in claim 5 , wherein the Kurtosis measurement is greater than 3.

8. The system as claimed in claim 5 , the one or more memory units storing instructions executable by the one or more hardware processors to perform operations further comprising:
evaluating a reduction in complexity for performing data privacy preservation by using the k-anonymity data sanitization model as compared to using the k-anonymity l-diversity data sanitization model.

9. A non-transitory computer-readable medium storing database privacy protection instructions executable by one or more hardware processors to perform operations comprising:
determining a distribution pattern of one or more database attributes by applying Kurtosis measurement of data corresponding to the attributes to ascertain whether the distribution pattern is leptokurtic;
determining an adversary information gain for a k-anonymity data sanitization model and adversary information gain for a k-anonymity l-diversity data sanitization model, wherein the adversary information gain is the difference between entropy of S and a conditional entropy H(S|Q), and wherein S corresponds to a set of the attributes;
comparing the adversary information gain of the k-anonymity data sanitization model with the adversary information gain of the k-anonymity l-diversity data sanitization model repeatedly until the adversary information gain of the k-anonymity data sanitization model equals the adversary information gain of the k-anonymity l-diversity data sanitization model;
determining an optimal value of l for performing l-diversity based data sanitization on database records related to the attributes and an optimal value of k for performing k-anonymity based data sanitization on the attributes; and
performing privacy preservation of the attributes by only k-anonymity data sanitization model when k is greater than l and the distribution pattern is leptokurtic.

10. The medium as claimed in claim 9 , storing instructions executable by one or more hardware processors to perform operations further comprising:
determining the optimal value of l for performing l-diversity based data sanitization and k for performing k-anonymity based data sanitization using a Normalized Certainty Penalty.

11. The medium as claimed in claim 9 , storing instructions executable by one or more hardware processors to perform operations further comprising:
determining a privacy disclosure probability that decreases with increase in value of k/l.

12. The medium as claimed in claim 9 , storing instructions executable by one or more hardware processors to perform operations further comprising:
evaluating a reduction in complexity for performing data privacy preservation by using the k-anonymity data sanitization model as compared to using the k-anonymity l-diversity data sanitization model.

Assignees

• Ukil Arijit
• Sen Jaydip
• Tata Consultancy Services Ltd

Inventors

• Ukil Arijit
• Sen Jaydip

Classifications

• G06F21/6227 Primary

  where protection concerns the structure of data, e.g. records, types, queries · CPC title

• G06F21/60 Primary

  Protecting data · CPC title

• G06F21/6254

  by anonymising data, e.g. decorrelating personal data from the owner's identification · CPC title