Privacy Data Containment via Deep Learning Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to adequately safeguard and report user privacy data, leading to potential compromise or unauthorized sharing of sensitive information, which can result in financial loss, credit issues, and user frustration.
Innovation Solution
A customizable privacy data system using deep learning models and modular containers for data scanning, classification, and reporting, ensuring compliance with regulations like GDPR, and enabling secure handling and storage of user data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If private user data is transmitted between devices and systems, then data communication and processing are enabled, but the risk of unauthorized sharing and data compromise increases
Solution Approach 1:
The patent segments private data into multiple partitions, where each partition contains only a subset of the complete private information. Multiple devices must collaborate to reconstruct the full data, preventing any single device from accessing complete private information. This resolves the contradiction by enabling data processing across devices while mitigating the risk of data compromise through distributed storage.
Solution Approach 2:
The patent introduces a trusted execution environment (TEE) as an intermediary layer between data storage and processing operations. The TEE acts as a secure mediator that enables data processing while preventing unauthorized access to raw private data. This resolves the contradiction by facilitating data communication and processing productivity while the TEE intermediary protects against data compromise risks.
2Reliability
If data is stored in multiple devices for distribution, then data security and compliance are improved, but system complexity increases
Solution Approach 1:
The patent implements a universal data partitioning framework that can be applied across different device types and storage configurations. The same partitioning and reconstruction mechanisms work across diverse systems, reducing the complexity burden despite distributed storage. This resolves the contradiction by maintaining high data security through multi-device storage while using a universal approach to manage system complexity.
Solution Approach 2:
The patent employs nested encryption layers where data is encrypted in multiple hierarchical levels before distribution across devices. Each device stores encrypted partitions with different encryption layers, and reconstruction requires combining multiple layers. This nested structure improves data security through layered protection while organizing the complexity in a manageable hierarchical manner.
3Measurement precision
If deep learning models are used for data classification, then classification accuracy is improved, but computational resource consumption increases
Solution Approach 1:
The patent performs data classification using deep learning models during the data ingestion and partitioning phase, before data is distributed to multiple devices. By completing the computationally intensive classification task upfront, the system achieves high classification accuracy while avoiding repeated computational energy consumption during data reconstruction and processing operations. This resolves the contradiction by performing accurate classification in advance when computational resources are more readily available.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Aspects of the present disclosure involve, a customizable system and infrastructure which can receive privacy data from varying data sources for privacy scanning, containment, and reporting. In one embodiment, data received is scanned for privacy data extraction using various data connectors and decryption techniques. In another embodiment, the data extracted is transferred to a privacy scanning container where the data is analyzed by various deep learning models for the correct classification of the data. In some instances, the data extracted may be unstructured data deriving form emails, case memos, surveys, social media posts, and the like. Once the data is classified, the data may be stored or contained according to the classification of the data. Still in another embodiment, the classified data may be retrieved by an analytics container for use in reporting.