ANONYMIZATION APPARATUS, ANONYMIZATION METHOD, AND COMPUTER PROGRAM

• US 20140317756A1
• Filed: 12/06/2012
• Published: 10/23/2014
• Est. Priority Date: 12/15/2011
• Status: Abandoned Application

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Takahashi, Tsubasa
• Application Number:
  US14/365,615
• Publication Number:
  US 20140317756A1
• Time in Patent Office: Days
• US Class Current: 726/26
• CPC Class Codes: G06F 21/60 Protecting data
• CPC Class Codes: G06F 21/6254 by anonymising data, e.g. decorrelating personal data from the owner's identification

Abstract

Provided is an anonymization apparatus for optimally and fully performing anonymization, in anonymizing history information, in a state where a specific property existing in a plurality of records with an identical identifier is maximally maintained.

This anonymization apparatus includes record extracting means for extracting, with respect to history information including a plurality of records each of which includes unique identification information associated with at least a quasi-identifier and sensitive information, on the basis of smallness of ambiguity of a property existing between the plurality of records that are able to satisfy desired anonymity and share a specific unique identifier, a record with other unique identifier different from the specific unique identifier from the history information and anonymizing means for giving commonality to and thereby abstracting the quasi-identifier each included in the plurality of records, so that an individual attribute in the plurality of records extracted by the record extracting means satisfies the desired anonymity.

Claims

1. An anonymization apparatus, comprising:

a record extracting unit that extracts, with respect to history information including a plurality of records each of which includes unique identification information associated with at least a quasi-identifier and sensitive information, on the basis of smallness of ambiguity of a property existing between the plurality of records that are able to satisfy desired anonymity and share a specific unique identifier, a record with other unique identifier different from the specific unique identifier from the history information; and

an anonymizing unit that gives commonality to and thereby abstracts the quasi-identifier each included in the plurality of records, so that an individual attribute in the plurality of records extracted by the record extracting unit satisfies the desired anonymity.

2. The anonymization apparatus of claim 1, wherein the record extracting unit extracts a record in which ambiguity which is able to be estimated after abstraction is the smallest, in the records having the other unique identifier, the records being abstracted with the record having the specific unique identifier.

3. The anonymization apparatus according to claim 1, wherein the record extracting unit arranges the record having the specific unique identifier and the extracted record having the record with the other unique identifier different from the extracted specific unique identifier into one group, and the anonymizing unit performs the abstraction by the group.

4. The anonymization apparatus according to claim 1, further comprising:

a property analysis unit that extracts the property from the history information by analyzing a plurality of records configuring the history information.

5. The anonymization apparatus according to claim 1, further comprising:

an evaluation unit that evaluates importance of the properties when a plurality of types of the properties exist, and selects an important property to be prioritized in the extraction by the record extracting unit.

6. The anonymization apparatus according to claim 5, further comprising:

a request reception unit that enables to entry of a request related to the property whose storage is desired in the plurality of types of the properties.

7. The anonymization apparatus according to claim 6, wherein the evaluation unit evaluates importance of the property extracted by the property analysis unit after enhancing the importance of the property, which is entered by the request reception unit, compared with the other property.

8. An anonymization method, comprising:

extracting, with respect to history information including a plurality of records each of which includes unique identification information associated with at least a quasi-identifier and sensitive information, on the basis of smallness of ambiguity of a property existing between the plurality of records that are able to satisfy desired anonymity and share a specific unique identifier, a record with other unique identifier different from the specific unique identifier from the history information by using a computer; and

giving commonality to and thereby abstracting the quasi-identifier each included in the extracted plurality of records, so that an individual attribute in the plurality of records satisfies the desired anonymity, by using the computer or a different computer.

9. The anonymization method according to claim 8, wherein the extracting comprises extracting a record in which the ambiguity which is able to be estimated after abstraction is the smallest, in the records having the unique identifier, the records being abstracted with the record having the specific unique identifier.

10. A non-transitory computer-readable media storing an computer program, causing a computer to execute:

a record extracting function for extracting, with respect to history information including a plurality of records each of which includes unique identification information associated with at least a quasi-identifier and sensitive information, on the basis of smallness of ambiguity of a property existing between the plurality of records that are able to satisfy desired anonymity and share a specific unique identifier, a record with other unique identifier different from the specific unique identifier from the history information; and

an anonymizing function for giving a commonality to and thereby abstracting the quasi-identifier each included in the plurality of records, so that an individual attribute in the plurality of records extracted by the record extracting function satisfies the desired anonymity.