Anonymized Data Generation via Dynamic Grouping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing anonymization methods fail to ensure adequate anonymity for data with multiple records from the same provider, particularly when these records are spatially clustered, leading to potential identification of the information provider, and compromise analysis accuracy due to large mesh sizes or restrictive grouping conditions.

Innovation Solution

A method that groups data blocks with specific frequency distribution patterns and replaces numeric attribute values within each group with calculated values, ensuring anonymity without relation to other areas, allowing for precise analysis while preventing identification of the information provider.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is anonymized by deleting IDs and using conventional mesh-based grouping, then anonymity is provided, but analysis accuracy deteriorates due to large mesh sizes or restrictive grouping conditions

Engineering Contradiction:
ImproveanonymityVSAvoidanalysis accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent segments the data processing into two distinct phases: first performing anonymization by deleting IDs and grouping records, then performing analysis on the anonymized data. This segmentation allows each phase to be optimized independently - the grouping phase prioritizes anonymity while the analysis phase can achieve high precision within those anonymous groups.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the grouping parameters dynamically based on data characteristics. Instead of using fixed mesh sizes, it adjusts grouping conditions to include sufficient records while maintaining spatial precision. This allows the system to achieve both anonymity (through adequate grouping) and analysis accuracy (through optimized group boundaries).

Inventive Principle:
Principle #35Parameter changes

2Reliability

If mesh size is increased to ensure anonymity, then anonymity is improved, but analysis accuracy deteriorates due to loss of spatial precision

Engineering Contradiction:
ImproveanonymityVSAvoidmesh size
Core Design Contradiction:
ReliabilityVSLength of stationary object

Solution Approach 1:

The patent introduces dynamic adjustment of grouping parameters based on the distribution and density of records. Rather than using static mesh sizes, the system adapts group boundaries and sizes to match actual data patterns, ensuring anonymity is maintained while preserving spatial precision where possible.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent applies different grouping strategies to different spatial regions based on local data characteristics. In areas with high record density, smaller groups can be formed maintaining anonymity; in sparse areas, larger groups are used to ensure sufficient records per group. This local adaptation optimizes both anonymity and spatial precision.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP2947597B1Confidential data generation method and device
Publication Date: 2019.06.19 FUJITSU LTD
  • EP2947597B1 patent drawingFigure 1~2
  • EP2947597B1 patent drawingFigure 3
  • EP2947597B1 patent drawingFigure 4

AI summary

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.