Selective Dataset Encryption Using Heat Map Relevancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Deep learning models stored on clouds are vulnerable to security attacks, and conventional homomorphic encryption techniques require complex operations, leading to unnecessary computations and power loss.
Innovation Solution
A method and system for selectively encrypting datasets using a common heat map to determine the relevance of data points, assigning different encryption levels based on their importance, and using homomorphic encryption keys with varying strengths to protect sensitive data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional homomorphic encryption techniques are used to protect user data, then data security is improved, but computational complexity increases and power loss occurs
Solution Approach 1:
The patent applies local quality by assigning different encryption strengths to different regions of the dataset based on their sensitivity levels. Sensitive regions (identified through heat map analysis) receive stronger encryption, while less sensitive regions use weaker encryption, thereby optimizing the balance between security and computational efficiency.
Solution Approach 2:
The dataset is segmented into multiple regions with different sensitivity levels using heat map analysis. This segmentation allows the system to apply differentiated encryption strategies to different parts of the data, avoiding the need to encrypt entire datasets with maximum strength and reducing overall computational overhead.
2Reliability
If stronger encryption keys are used to protect data, then data security is improved, but unnecessary computations increase and operational delays occur
Solution Approach 1:
Different encryption key strengths are applied to different data regions based on their sensitivity. High-sensitivity regions use strong encryption keys, while low-sensitivity regions use weaker keys, eliminating unnecessary strong encryption computations and improving operational speed.
Solution Approach 2:
The encryption key strength parameter is dynamically adjusted based on the sensitivity level of each data region. This parameter change allows the system to optimize the balance between security and speed by using appropriate encryption strength for each specific data region rather than applying uniform strong encryption throughout.
3Productivity
If selective encryption is implemented based on data sensitivity, then computational efficiency is improved, but system complexity increases
Solution Approach 1:
The system performs preliminary heat map analysis to identify and classify sensitive regions before the encryption process. This preliminary action creates a sensitivity map that guides the selective encryption process, allowing the system to efficiently determine which regions require encryption and what strength is needed, thereby managing complexity through pre-processing.
Solution Approach 2:
The heat map analysis acts as an intermediary between the raw data and the encryption process. It provides sensitivity information that mediates the decision-making for encryption strength, simplifying the overall system architecture by decoupling the sensitivity assessment from the encryption execution.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and a system of selective encryption of a test dataset is disclosed. In an embodiment, the method may include determining a relevancy grade associated with each of a plurality of datapoints within a test dataset by comparing the test dataset with a common heat map, wherein the common heat map may be generated using a plurality of training datasets. The method may further include calculating, based on the relevancy grade, an encryption level associated with each of the plurality of datapoints. The method may further include selectively encrypting at least one datapoint from the plurality of datapoints based on the encryption level associated with each of the plurality of datapoints. The at least one data point may be rendered to a user after being decrypted.