Privacy Knowledge Base for Balancing Data Utility and Anonymization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing data anonymization methods face challenges in balancing privacy protection and data utility, as excessive anonymization reduces the usefulness of data for analysis, and configuring privacy models to achieve this balance is complex and requires deep knowledge of algorithms and parameters.

Innovation Solution

A method is developed to create a Privacy Knowledge Base that determines unique combinations of anonymization protocols and parameters for relevant data fields, calculating privacy and data utility scores to optimize privacy settings for each analysis function, thereby automating the selection of optimal privacy settings and balancing privacy and data utility.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If more data is anonymised to reduce privacy risk, then privacy protection is improved, but data utility deteriorates

Engineering Contradiction:
Improveprivacy protectionVSAvoiddata utility
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The patent applies parameter changes by adjusting anonymization protocol parameters (such as k-anonymity level, generalization degree, suppression ratio) to achieve optimal balance between privacy protection and data utility. The system evaluates different parameter combinations and selects those that maximize privacy while preserving analytical value for cybersecurity analysis.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent implements dynamic anonymization where the degree of anonymization is adjusted based on the specific analysis function requirements. Different analysis functions receive different levels of anonymization - more sensitive analyses get higher privacy protection while less sensitive ones retain more utility, making the anonymization process adaptive rather than static.

Inventive Principle:
Principle #15Dynamics

2Reliability

If anonymization protocols are applied to protect privacy, then privacy risk is reduced, but configuration complexity increases

Engineering Contradiction:
Improveprivacy risk reductionVSAvoidconfiguration complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements self-service through automated anonymization configuration. The system automatically selects appropriate anonymization protocols and parameters based on the analysis function requirements and data characteristics, eliminating the need for manual configuration by privacy experts. The automated system evaluates privacy risks and adjusts anonymization settings without human intervention.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent uses feedback mechanisms where the system continuously evaluates the effectiveness of anonymization settings by monitoring privacy risk metrics and data utility outcomes. Based on this feedback, the system automatically adjusts anonymization parameters to optimize the balance between privacy protection and analytical value, creating a closed-loop control system.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If deep knowledge of algorithms and parameters is required to configure privacy models, then privacy protection accuracy is improved, but ease of operation deteriorates

Engineering Contradiction:
Improveprivacy protection accuracyVSAvoidconfiguration ease
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent introduces an intermediary automated system that mediates between the user's simple requirements and the complex anonymization algorithms. The system translates high-level privacy requirements into specific algorithm configurations automatically, acting as an intermediary that handles the computational complexity while presenting a simple interface to users.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces manual mechanical configuration of anonymization parameters with automated computational systems. Instead of requiring users to manually tune algorithm parameters based on deep knowledge, the system uses automated evaluation and optimization algorithms to determine the best settings, substituting human expertise with computational intelligence.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4398140A1Privacy knowledge base
Publication Date: 2024.07.10 BRITISH TELECOM PLC
  • EP4398140A1 patent drawingFigure 1
  • EP4398140A1 patent drawingFigure 2
  • EP4398140A1 patent drawingFigure 3

AI summary

A computer-implemented method is provided for use in anonymizing data, the data having a plurality of data attributes, or fields. The method is performed by a computer system 501 configured to provide anonymized data to a plurality of different analysis functions 515, 517, 519, 521 for analyzing the data, the data being anonymized by applying one or more of a plurality of predetermined anonymization protocols, each protocol having an anonymization protocol parameter that can be specified in order to determine the degree of anonymization applied to the data. The method comprises determining, for each analysis function: privacy settings, each comprising a unique combination of anonymization protocol applied to the data, anonymization protocol parameter value and data attribute; a privacy score P for each of the determined unique combinations of anonymization protocol applied to the data, anonymization protocol parameter value and data attribute; and a data utility score U for each of the determined unique combinations of anonymization protocol applied to the data, anonymization protocol parameter value and data attribute. The method further comprises processing the results, for each analysis function, by: ranking the unique combinations of anonymization protocol, anonymization protocol parameter and data attribute based on the privacy score and/or data utility score; and storing the ranked results for subsequent selection and use by the computer system to process an anonymized set of data, anonymized according to an identified anonymization protocol, according to the analysis function. A corresponding computer system and computer program are also provided.