Confidential Data Anonymization via Segmented Encryption and Statistical Smoothing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a challenge in ensuring the confidentiality and secure usage of sensitive data, such as salary information, in computer systems, where users are hesitant to share due to privacy concerns, and existing technologies struggle to balance data quality and coverage while maintaining user anonymity.
Innovation Solution
A system is developed that securely collects and processes confidential data by using a confidential data frontend and backend architecture, employing encryption techniques, anonymization, and machine learning models to ensure data privacy and provide accurate insights, while incentivizing users to contribute data through 'give-to-get' incentives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If confidential data is collected and stored in a centralized database, then statistical insights and coverage are improved, but user privacy and data security are compromised
Solution Approach 1:
The patent segments confidential data into multiple encrypted components stored in separate tables (submission tables with different encryption keys). This allows statistical analysis to be performed on aggregated data while individual records remain protected through cryptographic segmentation, resolving the contradiction between data utility and privacy protection.
Solution Approach 2:
The patent introduces cryptographic intermediaries (encryption keys, hashing functions) that mediate between raw confidential data and statistical analysis operations. These intermediaries enable insights to be derived without exposing the actual confidential values, balancing information availability with privacy preservation.
2Adaptability or versatility
If data grouping is increased to improve coverage, then more insights are provided, but the number of samples per grouping decreases reducing data quality
Solution Approach 1:
The patent adds the dimension of cryptographic transformation to the data analysis space. By operating on encrypted representations rather than raw values, the system can create multiple grouping dimensions without being constrained by sample size, as the cryptographic layer preserves statistical properties across different grouping granularities.
3Reliability
If encryption is applied to protect confidential data, then security is improved, but data processing and analysis become more complex
Solution Approach 1:
The patent applies preliminary cryptographic transformations (encryption, hashing) during data ingestion and storage. By pre-processing data into secure representations, the system eliminates the need for complex cryptographic operations during analysis queries, as the encrypted structure itself enables statistical operations without decryption.
Data Source
AI summary
In an example, an anonymized set of confidential data data values of a first confidential data type is obtained. Then an anonymized set of confidential data data values of a second confidential data type is also obtained. A multiplier following a log-normal distribution is determined for the anonymized set of confidential data data values of the first confidential data type. Then smoothing is performed independently for the anonymized set of confidential data data values of the first confidential data type and the multiplier. Percentiles for the anonymized set of confidential data data values of the second confidential data type are then determined using the smoothed anonymized set of confidential data data values of the first confidential data type and the smoothed multiplier.