Local Model Selection for Secure User Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing user identification and personal data analysis methods are vulnerable to unauthorized access, as they involve transferring personal data over networks, compromising information security.
Innovation Solution
A system and method for selecting a model to describe a user based on their behavior data, using a base model and correcting models to enhance accuracy without transferring personal data, ensuring data security by retraining models locally on user devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If personal data is transferred over networks for user identification and analysis, then user identification accuracy is improved, but information security deteriorates
Solution Approach 1:
The patent extracts personal data from the data processing system and stores it locally in a personal data storage unit within the user's computing device. Only non-personal processed data is transferred over the network, eliminating the security risk of personal data transmission while maintaining identification functionality.
Solution Approach 2:
The patent introduces a personal data storage unit as an intermediary component that resides locally on the user's device. This intermediary stores personal data securely and provides processed non-personal data to external systems, acting as a buffer that prevents direct exposure of personal information while enabling identification services.
2Object-affected harmful factors
If personal data is stored locally on user devices, then information security is improved, but network data transfer capabilities deteriorate
Solution Approach 1:
The patent segments data into two categories: personal data stored locally in the personal data storage unit, and non-personal processed data transmitted over the network. This segmentation allows secure local retention of sensitive information while enabling necessary data exchange for service functionality.
3Adaptability or versatility
If trained models are used to predict user preferences, then service personalization is improved, but computational resource demands deteriorate
Solution Approach 1:
The patent performs model training and data processing in advance, storing the results in the personal data storage unit. When providing services, the system retrieves pre-processed non-personal data and applies pre-trained models, avoiding the need for real-time training and reducing computational resource demands during service delivery.
Data Source
AI summary
Disclosed herein are systems and methods for selection of a model to describe a user. In one aspect, an exemplary method comprises, creating data on preferences of the user based on previously gathered data on usage of a computing device by the user and a base model that describes the user, wherein the base model is previously selected from a database of models including a plurality of models, determining an accuracy of the data created on the preferences of the user, wherein the determination is based on observed behaviors of the user, when the accuracy of the data is determined as being less than a predetermined threshold value, selecting a correcting model related to the base model, and retraining the base model, and when the accuracy of the data is determined as being greater than or equal to the predetermined threshold value, selecting the base model to describe the user.