Contextualizing Attacker Knowledge for Data Anonymization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data anonymization methods, such as k-anonymity, face challenges in reliably capturing quasi-identifiers (QIDs) to effectively anonymize datasets, leading to potential overestimation of attacker knowledge and poor quality anonymized data.
Innovation Solution
A computer-implemented system that contextualizes the attacker's background knowledge by matching target attributes with data consumer background knowledge, using natural language processing and similarity measures, to identify and model QIDs accurately, thereby improving the quality of anonymization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional k-anonymity methods are used to anonymize datasets, then data security is improved, but the quality of anonymized data deteriorates due to poor capture of quasi-identifiers and overestimation of attacker knowledge
Solution Approach 1:
The system performs preliminary contextualization of attacker background knowledge before applying anonymization. By matching target attributes with background knowledge attributes using NLP and similarity measures, the system pre-identifies accurate quasi-identifiers and models attacker knowledge capabilities in advance, preventing overestimation and enabling more precise anonymization that preserves data quality
Solution Approach 2:
The system dynamically adjusts the anonymization parameters based on the contextualized attacker knowledge model. By changing the parameters of quasi-identifier selection and anonymization intensity according to the modeled attacker capabilities, the system achieves optimal balance between security and data quality, avoiding both over-anonymization and under-anonymization
2Reliability
If comprehensive attacker background knowledge is assumed to ensure security, then data security is improved, but data quality deteriorates due to over-anonymization
Solution Approach 1:
The system applies different levels of anonymization to different attributes based on the contextualized attacker knowledge model. By identifying which attributes are actually relevant to attacker knowledge through NLP matching and similarity measures, the system applies targeted anonymization only where necessary, preserving quality in non-critical attributes while maintaining security in sensitive areas
Solution Approach 2:
The system performs partial contextualization of attacker knowledge by matching only the most relevant target attributes with background knowledge attributes. This selective approach avoids the excessive assumption of complete attacker knowledge, enabling the system to achieve adequate security with less aggressive anonymization, thereby preserving data quality
Data Source
AI summary
A system and related method for facilitating data anonymization. The system may include a contextualizer (CTX) configured to match, in a matching operation, target attributes of the target dataset (TD) with one or more attributes of data representing a data consumer (DC)'s background knowledge (BK) for the target dataset (TD). As a result of the matching operation, a contextualized data consumer (DC)'s background knowledge is generated, which is representative of the data consumer (DC)'s background knowledge relative to the target dataset. An output interface (OUT) of the system (SYS) provides the contextualized data consumer (DC)'s background knowledge data to an anonymizer (AN) for anonymizing the target dataset.