Secure Database Group-By Using One-Hot Encoding and MPC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing secure multi-party computation (MPC) protocols for group-by operations in databases face high computational and communication complexities due to the need for ciphertext sorting and numerous comparison operations, especially when data is distributed across multiple databases controlled by different entities with varying data privacy requirements.
Innovation Solution
Perform local operations on 'group-by' attributes using one-hot encoding to convert categorical data into numerical data, and utilize oblivious shuffling protocols for sorting, reducing computational complexity while ensuring data privacy through secure multi-party computation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional secure MPC protocols are used for group-by operations, then data privacy is protected, but computational complexity and communication overhead increase significantly
Solution Approach 1:
The patent applies preliminary action by performing local one-hot encoding on categorical attributes before the secure MPC computation. This preprocessing step converts categorical data into numerical format in advance, eliminating the need for complex ciphertext sorting operations during the actual group-by computation, thus reducing computational complexity while maintaining privacy protection
Solution Approach 2:
The patent extracts and processes only the necessary attributes locally before secure computation. By taking out the categorical attributes and converting them to numerical form beforehand, the system avoids the need to perform expensive cryptographic operations on all attributes, thereby reducing both computational complexity and communication overhead
2Ease of operation
If ciphertext sorting is performed in secure MPC, then group-by operations can be executed, but communication complexity increases due to numerous comparison operations
Solution Approach 1:
The patent substitutes the mechanical ciphertext sorting process with a mathematical approach. By converting categorical attributes to numerical form via one-hot encoding before computation, the system replaces the need for iterative ciphertext comparison and sorting operations with direct numerical aggregation, significantly reducing communication volume
3Adaptability or versatility
If data is distributed across multiple databases with different privacy requirements, then data collaboration is enabled, but computational efficiency decreases due to secure MPC overhead
Solution Approach 1:
The patent applies segmentation by dividing the computation into two distinct phases: local preprocessing (one-hot encoding) performed by individual data holders, and centralized secure MPC computation performed by the trusted third party. This segmentation allows each party to perform only necessary operations locally, reducing the overall computational burden on the secure MPC system while enabling data collaboration across multiple databases
Data Source
AI summary
The present disclosure involves methods, apparatus, and systems for processing group-by operations in databases. In one example, a method includes receiving a query to perform a group-by operation based on first attributes and second attributes of a plurality of entities. The first attributes are private data owned by a first participant of a secure multi-party computation (MPC), and the second attributes are private data owned by a second participant of the secure MPC. The method further includes mapping the first attributes to a first set of numbers, and calculating, based on secure MPC, a secret share of a dataset. The dataset includes the first set of numbers and the second attributes. The method further includes performing the group-by operation including arranging the plurality of entities into groups based on at least the first set of numbers, and aggregating the second attributes of entities in each of the groups.