Dynamic Anonymization of Automotive Subscription Data via PII Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data anonymization methods fail to adequately balance data privacy and security concerns with business value, as Personally Identifying Information (PII) remains stored, making data susceptible to breaches.
Innovation Solution
A system comprising a determination component to identify PII elements based on an anonymization criterion and a storage component to store non-identifying data elements in a non-identifying data store, utilizing techniques such as chunking, rounding, and artificial intelligence to ensure data anonymization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If all data is stored in a data store to maintain business value for data analytics, then data utility is improved, but data security deteriorates because Personally Identifying Information remains stored and susceptible to data breaches
Solution Approach 1:
The patent segments data into Personally Identifying Information (PII) and non-PII components using an anonymization criterion. The determination component divides the data set by evaluating each data element against the criterion, separating identifiable from non-identifiable information. This segmentation allows the system to store only non-PII data in the data store, maintaining data utility for analytics while eliminating PII-related security risks.
Solution Approach 2:
The patent extracts and removes PII elements from the data set before storage. The determination component identifies and extracts PII data elements based on the anonymization criterion, and the storage component stores only the remaining non-identifying data elements. This extraction process eliminates the security vulnerability while preserving the analytical value of the anonymized data.
2Loss of information
If PII is stored to maintain complete data for analytics, then data completeness is improved, but privacy protection deteriorates
Solution Approach 1:
The system segments data completeness into two components: PII elements and non-PII elements. By separating these components through the anonymization criterion evaluation, the system maintains completeness of analytical data (non-PII) while deliberately excluding PII elements to ensure privacy protection. The determination component performs this segmentation by comparing each data element against the anonymization criterion.
Solution Approach 2:
The system extracts PII elements from the complete data set using the determination component and anonymization criterion. The storage component then stores only the extracted non-PII elements, achieving a balance where data completeness for analytics is maintained through retention of non-identifying elements, while privacy protection is ensured through removal of identifying elements.
3Reliability
If data anonymization is applied to protect privacy, then privacy protection is improved, but data queryability deteriorates because not all data can be queried
Solution Approach 1:
The patent applies dynamic anonymization where the anonymization criterion and its parameters can be adjusted based on query requirements and privacy needs. The determination component evaluates data elements against the anonymization criterion, which can be dynamically configured. This dynamic approach allows the system to balance privacy protection with queryability by adjusting the strictness of anonymization based on the specific analytical context.
Solution Approach 2:
The system changes parameters of the anonymization criterion to optimize the balance between privacy protection and data queryability. The determination component uses configurable parameters within the anonymization criterion to control the degree of anonymization. By adjusting these parameters, the system can maintain higher queryability when privacy risks are lower, and apply stricter anonymization when privacy protection is paramount.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method can comprise determining, by a device comprising a processor, personally identifying data elements of data, representative of a group of transactions, that comprise personally identifying information according to an anonymization criterion associated with personally identifying information being determined not to be satisfied by the data elements, and storing, by the device, non-identifying data elements of the data to a non-identifying data store.