Data Anonymization via Risk-Utility Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data anonymization techniques face challenges in balancing data privacy protection with the utility of data for analysis, as existing methods often compromise data accuracy and utility, making it difficult to determine the optimal level of anonymization for different data elements and analysis goals.
Innovation Solution
A two-step approach is proposed, where data owners specify protection goals and importance values for data fields, allowing for a risk and utility score computation to determine the appropriate anonymization technique, such as k-anonymity or differential privacy, to ensure data privacy while maintaining analysis accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data anonymization techniques are applied to protect privacy, then re-identification risk is reduced, but data utility and analysis accuracy deteriorate
Solution Approach 1:
The system dynamically adjusts anonymization parameters based on data sensitivity classification and analysis requirements. Different anonymization techniques (k-anonymity, l-diversity, t-closeness, differential privacy) are selectively applied to different data fields based on their sensitivity level, allowing the system to optimize the balance between privacy protection and data utility rather than applying a fixed anonymization level to all data
Solution Approach 2:
The patent implements field-level anonymization where different anonymization techniques and parameters are applied to different data fields based on their sensitivity classification. Critical fields receive stronger anonymization while less sensitive fields maintain higher utility, resolving the contradiction by making anonymization quality local rather than uniform across the entire dataset
2Reliability
If strong anonymization techniques are applied, then privacy protection is improved, but data accuracy for analysis deteriorates
Solution Approach 1:
The system changes anonymization parameters (k-value, l-value, t-value, epsilon for differential privacy) based on data sensitivity classification and analysis requirements. By adjusting these parameters, the system can optimize the trade-off between privacy protection strength and data accuracy for different use cases, rather than using fixed strong anonymization for all data
3Adaptability or versatility
If data fields are classified by sensitivity and importance, then optimal anonymization can be determined, but system complexity increases
Solution Approach 1:
The patent segments data fields into different sensitivity levels (critical, high, medium, low) and importance categories (identifier, quasi-identifier, non-identifier). This segmentation allows the system to apply appropriate anonymization techniques to each segment, optimizing privacy protection while managing complexity through structured classification rather than uniform processing of all fields
Data Source
AI summary
A set of data is received for a data analysis. The set of data includes personal identifiable information. The set of data is anonymized to protect the privacy information. Risk rates and utility rates are determined for a number of combinations of anonymization techniques defined correspondingly for data fields from the set of data. A risk rate is related to a privacy protection failure when defining first anonymized data through applying a combination of anonymization techniques for the data fields. A utility rate is related to accuracy of the data analysis when applied over the anonymized data. Based on evaluation of the risk rates and the utility rates, one or more anonymization techniques from the number of anonymization techniques are determined. The set of data is anonymized according to a determined anonymization techniques and/or a combination thereof.