Application Execution Control Using Predicted User Intent
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic devices often execute services or functions contrary to the intention of a legitimate user, causing inconvenience.
Innovation Solution
An electronic device employs a machine learning model to predict the probability of use based on historical data, allowing it to control the execution of applications and support elements accordingly, with user interface confirmation and automatic control options.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the electronic device executes services based on user requests, then the service functionality is provided, but services may be executed contrary to the user's intention
Solution Approach 1:
The system performs preliminary actions by collecting historical usage data and training a machine learning model before actual service execution. The model predicts user intention in advance, allowing the device to anticipate whether a requested service aligns with the user's true intention before executing it, thus preventing unintended service execution.
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring actual user behavior and service execution outcomes. This feedback is used to retrain and refine the machine learning model, improving its ability to predict user intention accurately over time, thereby enhancing the alignment between service execution and user intention.
2Measurement precision
If the device uses machine learning models to predict user intention, then service execution accuracy improves, but device complexity increases
Solution Approach 1:
The system segments the intention prediction function into a separate machine learning model component, distinct from the core service execution logic. This modular approach allows the ML model to be trained and updated independently without affecting the overall system architecture, managing complexity while maintaining prediction accuracy.
Solution Approach 2:
The machine learning model acts as an intermediary between user service requests and the actual service execution. It processes historical data and predicts user intention, providing a decision layer that mediates whether to execute a requested service, thus improving accuracy without directly complicating the service delivery mechanism.
3Measurement precision
If the device collects historical usage data for model training, then prediction accuracy improves, but data processing time increases
Solution Approach 1:
The system performs preliminary data processing by continuously collecting and preprocessing historical usage data in the background during normal operation. This preprocessed data is ready for model training when needed, reducing the time required for actual prediction tasks while maintaining high prediction accuracy through comprehensive historical analysis.
Solution Approach 2:
The system implements periodic action by training the machine learning model at scheduled intervals rather than continuously. Historical data is collected continuously but processed in periodic batches, balancing the need for accurate predictions with the constraint of data processing time and system resource utilization.
Data Source
AI summary
An operating method of an electronic device includes receiving a request to control of at least one application that is providable by the electronic device, obtaining second information according to a machine learning model based on first information corresponding to a history of use regarding the at least one application and at least one support element. The machine learning model predicts at least a probability of use of the at least one support element which supports execution of the at least one application based on the first information, and the method executes a function corresponding to the request to control the at least one application to be provided by the electronic device, based on the second information including at least the probability of use of the at least one support element which supports the execution of the at least one application.