Salience Score User Model via Physiological Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for organizing, ranking, and retrieving documents do not effectively associate user salience or focus with content, leading to suboptimal filtering, ranking, and clustering based on user interest.
Innovation Solution
A system that uses physiological sensors and eye tracking data to determine a salience score for content, which is then used to update a user model and inform document organization, filtering, and retrieval processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional search engines organize content based on popularity and search terms, then content can be retrieved using simple algorithms, but the system cannot effectively capture user interest or salience
Solution Approach 1:
The patent introduces physiological sensors as an intermediary device that indirectly measures user interest through physiological responses (heart rate, skin conductance, pupil dilation) rather than directly analyzing user intent. This mediator translates complex user interest states into quantifiable signals that the system can process, resolving the contradiction between measurement precision and system complexity
Solution Approach 2:
The patent replaces traditional mechanical interaction-based user modeling (clicks, scrolls, search queries) with physiological signal-based measurement. By substituting the mechanical interaction system with a physiological sensing system, the patent achieves more precise measurement of user interest while maintaining system complexity at an acceptable level through automated signal processing
2Loss of information
If systems use semantic processing to organize content, then interrelationship of words can be captured, but user focus and salience remain unassociated with content
Solution Approach 1:
The patent merges semantic processing with physiological signal processing in an integrated system. The controller combines text analysis results with physiological response data to create a comprehensive user model that captures both the semantic meaning of content and the user's actual focus, preventing loss of user focus information while managing processing complexity through unified architecture
Solution Approach 2:
The patent creates a multi-functional system that simultaneously performs semantic analysis, physiological signal processing, and user model updating. This universal system handles multiple functions (content understanding, user state detection, personalized organization) in an integrated manner, reducing overall processing complexity compared to separate systems
3Reliability
If content is organized based on word popularity, then simple frequency counting can be used, but the system cannot determine actual user salience or engagement
Solution Approach 1:
The patent implements a feedback loop where physiological sensor data continuously updates the user model, which in turn refines content organization and presentation. This feedback mechanism progressively improves salience determination accuracy by incorporating real-time user physiological responses, while the automated nature of the feedback loop maintains operational simplicity
Data Source
AI summary
A method of using a user model with salience includes providing content to a user through a user interface. The method also includes receiving physiological data corresponding to a physiological response of the user as the user interacts with the content. The method may also include determining a salience score for the content based at least in part on the physiological response of the user, and updating a user model with the salience score.