Characteristic-Based Data Collection Across Devices for Private AI Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI systems face challenges in efficiently collecting training data while protecting personal information and ensuring the accuracy of generated AI models, particularly when relying on cloud or server-based data collection.
Innovation Solution
An electronic device equipped with a communication interface, memory, and processor that obtains data and characteristic information, selectively requests and receives data from external devices based on specific criteria, and generates an AI model using the collected data to maintain a balanced characteristic information profile.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If training data is collected from cloud or server, then data quantity can be increased, but personal information protection is compromised
Solution Approach 1:
The patent introduces an intermediary system that mediates between data sources and AI model training. This intermediary collects data from multiple devices, performs federated learning locally, and only transmits model updates rather than raw personal data, thus increasing training data quantity while protecting personal information privacy
Solution Approach 2:
The patent segments the centralised data collection process into distributed local learning units across multiple devices. Each device performs local training independently and contributes only aggregated model parameters to the central system, preventing exposure of individual personal information while accumulating diverse training data
2Measurement precision
If diverse training data is collected from multiple devices, then AI model accuracy is improved, but data collection complexity increases
Solution Approach 1:
The patent implements a universal data collection framework that can operate across multiple device types and communication protocols. The system uses standardized interfaces and common data formats that work universally across different devices, reducing collection complexity while enabling diverse data sources to improve AI model accuracy
3Manufacturing precision
If data is selectively collected based on characteristic information, then training data quality is improved, but data processing time increases
Solution Approach 1:
The patent performs preliminary analysis of data characteristic information before full data collection. The system pre-evaluates data sources based on their relevance to the AI model training objectives, pre-selects appropriate data subsets, and prepares data pipelines in advance, thereby improving training data quality while minimizing processing time delays
Data Source
AI summary
An electronic device includes a communication interface, a memory storing one or more instructions, and a processor configured to execute the one or more instructions stored in the memory. The processor is configured to execute the one or more instructions to obtain first data and characteristic information of the first data, control the communication interface to transmit a data request to an external device and receive characteristic information of second data from the external device, control the communication interface to receive the second data from the external device, based on the characteristic information of the first data and the characteristic information of the second data, determine training data including at least a portion of the first data and at least a portion of the second data, and generate the AI model, based on the determined training data.