Anonymized data generation method and apparatus

• US 9,747,467 B2
• Filed: 06/25/2015
• Issued: 08/29/2017
• Est. Priority Date: 01/16/2013
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Yamaoka, Yuji
• Primary Examiner(s): Hoffman, Brandon
• Assistant Examiner(s): Nipa, Wasika
• Application Number:
  US14/749,761
• Publication Number:
  US 20150294121A1
• Time in Patent Office: 796 Days
• CPC Class Codes: G06F 16/254 Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 21/6254 by anonymising data, e.g. decorrelating personal data from the owner's identification
• CPC Class Codes: H04L 63/04 for providing a confidential data exchange among entities communicating through data packet networks
• CPC Class Codes: H04L 63/0421 Anonymous communication, i.e. the party's identifiers are hidden from the other party or parties, e.g. using an anonymizer

Abstract

A method for generating anonymized data includes: (A) extracting, from plural data blocks, each of which includes a secret attribute value and a numeric attribute value, plural groups of data blocks, wherein each of the plural groups includes data blocks that include a first data block, which has not been grouped, whose frequency distribution of the secret attribute value satisfies a predetermined condition and whose numeric attribute values are within a certain area that has a predetermined size; and (B) replacing the numeric attribute values of the data blocks that belong to each group of the plural groups with a numeric attribute value calculated for the group. And, the certain area is determined without any relation with other certain areas for other groups.

Claims

1. An anonymized data generation method, comprising:

generating, by using a computer and from a plurality of data blocks each of which includes a secret attribute value and a numeric attribute value, a plurality of groups each of which includes plural data blocks that satisfy a predetermined condition, wherein the predetermined condition includes a first condition and a second condition, the first condition being a condition that frequency distribution of secret attribute values included in the plural data blocks matches a predetermined pattern, and the second condition being a condition that points represented by numeric attribute values included in the plural data blocks are included in a first area that has a predetermined size; and

replacing, by using the computer and for each of the plurality of groups, a numeric attribute value included in each of plural data blocks that belong to the group with a numeric attribute value calculated for the group, and

wherein the generating comprises;

classifying each of the plurality of data blocks into any of a plurality of second areas that have the predetermined size and do not overlap with each other, based on a numeric attribute value included in the data block; and

upon detecting that a set of data blocks included in a certain second area does not satisfy the predetermined condition, changing the set by deleting a data block from the set and/or adding, to the set, a data block included in another second area adjacent to the certain second area, so that changed set satisfies the predetermined condition.

2. The anonymized data generation method as set forth in claim 1, further comprising deleting, by using the computer and for each of the plurality of groups, secret attribute values included in plural data blocks that belong to group.

3. The anonymized data generation method as set forth in claim 1, wherein the generating comprises extracting, from a group of the plurality of groups, a first data block that is other than data blocks, which are mandatory for a state where the first condition is satisfied.

4. The anonymized data generation method as set forth in claim 1, wherein the first condition includes a lower limit value for a number of kinds of secret attribute values, and the changing comprises:

extracting, on a basis of a data block included in the set and from the another second area, a second data block that is to be added to the set so that a number of kinds of secret attribute values included in the data blocks included in the set and the second data block is equal to or greater than the lower limit value; and

determining the first area that has the predetermined size based on a numeric attribute value included in the extracted second data block.

5. The anonymized data generation method as set forth in claim 1, further comprising upon detecting that a third data block that does not belong to any of the plurality of groups, classifying, by using the computer, the third data block to a group of the plurality of groups, when a distance between a point represented by a numerical attribute value included in the third data block and a reference position of a certain area that includes points represented by numerical attribute values included in plural data blocks that belong to the group is equal to or less than a distance that corresponds to the predetermined size, and the first condition is still satisfied even when the third data block is added to the group.

6. The anonymized data generation method as set forth in claim 1, wherein the replacing comprises:

randomly generating, for each of the plurality of groups, an area that has the predetermined size and includes points represented by numeric attribute values included in plural data blocks which belong to the group; and

replacing, for each of the plurality of groups, a numeric attribute value included in each of plural data blocks which belong to the group with a numeric attribute value that corresponds to a position within an area generated for the group.

7. A non-transitory computer-readable storage medium storing a program that causes a computer to execute a process, the process comprising:

generating, from a plurality of data blocks each of which includes a secret attribute value and a numeric attribute value, plural a plurality of groups each of which includes plural data blocks that satisfy a predetermined condition, wherein the predetermined condition includes a first condition and a second condition, the first condition being a condition that frequency distribution of secret attribute values included in the plural data blocks matches a predetermined pattern, and the second condition being a condition that points represented by numeric attribute values included in the plural data blocks are included in a first area that has a predetermined size; and

replacing, for each of the plurality of groups, a numeric attribute value included in each of plural data blocks that belong to the group with a numeric attribute value calculated for the group, and

wherein the generating comprises;

classifying each of the plurality of data blocks into any of a plurality of second areas that have the predetermined size and do not overlap with each other, based on a numeric attribute value included in the data block; and

upon detecting that a set of data blocks included in a certain second area does not satisfy the predetermined condition, changing the set by deleting a data block from the set and/or adding, to the set, a data block included in another second area adjacent to the certain second area, so that changed set satisfies the predetermined condition.

8. An information processing apparatus, comprising

a memory;

and

a processor coupled to the memory and configured to;

generate, from a plurality of data blocks each of which includes a secret attribute value and a numeric attribute value, a plurality of groups each of which includes plural data blocks that satisfy a predetermined condition, wherein the predetermined condition includes a first condition and a second condition, the first condition being a condition that frequency distribution of secret attribute values included in the plural data blocks matches a predetermined pattern, and the second condition being a condition that points represented by numeric attribute values included in the plural data blocks are included in a first area that has a predetermined size; and

replace, for each of the plurality of groups, a numeric attribute value included in each of plural data blocks that belong to the group with a numeric attribute value calculated for the group, and

wherein the generating comprises;

classifying each of the plurality of data blocks into any of a plurality of second areas that have the predetermined size and do not overlap with each other, based on a numeric attribute value included in the data block; and

upon detecting that a set of data blocks included in a certain second area does not satisfy the predetermined condition, changing the set by deleting a data block from the set and/or adding, to the set, a data block included in another second area adjacent to the certain second area, so that changed set satisfies the predetermined condition.