Data Encoding for Secure Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for protecting sensitive data exchanged between a service user and a service provider are inadequate, as anonymization is insufficient for accurate analytics, and encryption is vulnerable to decryption by eavesdroppers, with the service provider being a weak link for data theft and unauthorized use.
Innovation Solution
A method involving encoding upload data to convert it into meaningless content, processing the encoded data, and decoding it back at the service user end, using an encoder module and decoder module to ensure only the service user can interpret the data, without affecting the processing outcome at the service provider.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is anonymized to protect privacy, then data security is improved, but analytics accuracy deteriorates
Solution Approach 1:
The patent introduces an intermediary encoding layer between the data and the analytics processing. The encoder converts readable data into encoded form that preserves statistical properties but eliminates identifiable information. This intermediary layer allows the service provider to process data without accessing its original meaning, thus securing privacy while maintaining analytics accuracy through the preserved statistical structure.
Solution Approach 2:
The patent transforms data parameters through encoding that changes the representation while preserving underlying statistical characteristics. The encoding modifies data values and structures in a way that prevents identification but maintains the relationships and patterns needed for accurate analytics, resolving the contradiction between privacy protection and analytical utility.
2Reliability
If data is encrypted to prevent eavesdropping, then data security is improved, but data availability for processing deteriorates
Solution Approach 1:
The encoding mechanism acts as an intermediary that enables data processing without requiring decryption. The encoded data maintains a structure that is directly processable by analytics tools, eliminating the need to decrypt data before processing. This resolves the contradiction by allowing data to remain protected while staying fully available for analytical operations.
3Productivity
If the service provider decrypts data for processing, then processing capability is improved, but vulnerability to data theft increases
Solution Approach 1:
The encoded data serves as an intermediary that enables the service provider to perform processing operations without accessing the original unencoded data. The encoding layer acts as a barrier that prevents data theft while allowing necessary analytics processing to occur on the encoded representations, thus maintaining productivity without increasing vulnerability.
Solution Approach 2:
The patent segments the data processing function into two independent parts: encoding/decoding operations performed by the service user, and analytics processing performed by the service provider on encoded data. This segmentation ensures that the service provider never handles decryptable data, preventing data theft while maintaining processing capability through the encoded data representations.
4Reliability
If comprehensive data fields are anonymized to prevent identification, then privacy protection is improved, but data utility for training algorithms deteriorates
Solution Approach 1:
The patent applies parameter changes through encoding that transforms data fields while preserving their statistical properties and relationships. The encoding modifies individual values and structures to prevent identification but maintains the patterns, distributions, and correlations needed for training algorithms, thus resolving the contradiction between privacy protection and data utility.
Solution Approach 2:
The encoding applies different transformation characteristics to different data fields based on their specific properties, preserving locally relevant statistical patterns while eliminating identifiable information. This localized approach ensures that each data field maintains its utility for training purposes while contributing to overall privacy protection.
Data Source
AI summary
The invention discloses a method of protecting data exchanged between a service user and a service provider, which method comprises the steps of encoding data by converting meaningful content of the data into meaningless content to obtain encoded upload data for sending to the service provider; processing the encoded upload data at the service provider to obtain encoded download data for sending to the service user; and decoding the encoded download data by converting meaningless content of the encoded download data into meaningful content of download data.