Privacy Data Reporting via Front Rear String Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data reporting methods face challenges in accurately matching and efficiently processing sensitive user data due to the inability to divide data into equal-length blocks, leading to low matching accuracy and high performance overheads.
Innovation Solution
A method where a terminal device divides data into front and rear string data based on a preset step, scrambles each part using a preset algorithm, and sends them to a server for matching in both directions, eliminating the need for conditional probability calculations and improving matching speed and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is divided into letter combinations of equal length for scrambling, then privacy protection is improved, but matching accuracy deteriorates for data that cannot be divided into equal length blocks
Solution Approach 1:
The patent divides to-be-reported data into front string data and rear string data based on a preset step, where the front string data length is an integer multiple of the preset step. This segmentation allows the front part to be divided into equal-length blocks for scrambling (maintaining privacy protection) while the rear part handles the remaining data (maintaining matching accuracy).
Solution Approach 2:
Different parts of the data are treated differently: the front string data is scrambled to protect privacy, while the rear string data is preserved to ensure accurate matching. This local differentiation resolves the contradiction between privacy protection and matching accuracy by applying different processing qualities to different segments of the same data structure.
2Reliability
If the server calculates conditional probabilities between scrambled letter combinations to obtain association relationships, then privacy protection is improved, but performance overheads increase
Solution Approach 1:
The patent extracts the conditional probability calculation from the server-side processing and relocates it to the terminal device. The terminal device pre-calculates association relationships between scrambled front string data and rear string data before reporting, eliminating the need for the server to perform these computationally intensive calculations, thus reducing performance overhead while maintaining privacy protection.
Solution Approach 2:
The terminal device performs preliminary processing by pre-calculating and storing association relationships between scrambled data segments before reporting to the server. This preliminary action shifts the computational burden from the server to the terminal, reducing the server's performance overhead while maintaining the privacy-protecting scrambling mechanism.
3Reliability
If data is scrambled to protect user privacy, then privacy protection is improved, but data matching complexity increases
Solution Approach 1:
By segmenting data into front string data (scrambled) and rear string data (unscrambled), the patent simplifies the matching process. The server can use the unscrambled rear string data for direct matching while only needing to verify the scrambled front string data structure, thereby reducing overall matching complexity compared to scrambling the entire data set.
Solution Approach 2:
The patent applies scrambling only to the front string data portion while leaving the rear string data in its original form. This local quality approach allows the matching system to work with unscrambled data for the majority of the matching operation, significantly reducing matching complexity while still providing privacy protection through the scrambled portion.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
Embodiments of this application disclose a privacy data reporting method and apparatus, and a storage medium. The method includes: A terminal device obtains to-be-reported data, divides the to-be-reported data into front string data and rear string data based on a preset step, separately scrambles the front string data and the rear string data, and sends the generated privacy data to a server. The server receives privacy data sent by a plurality of terminal devices, determines N pieces of first front string data, then obtains M pieces of to-be-reported data based on a plurality of pieces of privacy data, and finally determines to-be-reported data whose reporting rate is greater than a first preset value as target to-be-reported data. To be specific, according to the method in the embodiments, the terminal device divides the to-be-reported data into two parts, and separately scrambles the two parts to generate the privacy data. The server matches the privacy data sent by the terminal device with the to-be-reported data in two directions, namely, a head direction and a tail direction. In this way, matching accuracy and a matching speed are high.