Affective Response Estimation for Token Instances
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current user preference modeling in digital experiences is inadequate for discerning specific user preferences, limiting personalized content optimization, as existing methods primarily provide broad insights and fail to analyze reactions to specific details within content.
Innovation Solution
The system utilizes affective computing with sensors and algorithms to identify 'token instances' of interest, measuring user responses to specific elements in digital media, allowing for the creation of detailed user models by comparing reactions to these instances over time, enabling personalized content creation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional user preference modeling methods are used, then broad user preferences can be identified, but specific user preferences for individual elements cannot be discerned
Solution Approach 1:
The patent segments the content into discrete token instances (individual words, phrases, or elements) and measures user affective response to each segment separately. This allows the system to identify specific user preferences for individual elements rather than providing only broad content-level preferences, thereby resolving the contradiction between measurement precision and information loss.
2Productivity
If affective computing sensors are used to measure user emotional state, then continuous monitoring is possible, but only overall content response is obtained, not specific element reactions
Solution Approach 1:
The patent introduces token instances as intermediary elements that link the continuous affective computing measurements to specific content elements. By associating affective responses with token instances, the system can maintain continuous monitoring capability while also identifying reactions to specific elements, thus resolving the contradiction between productivity and measurement precision.
3Adaptability or versatility
If detailed analysis of user reactions to specific content elements is implemented, then personalized content optimization is improved, but system complexity increases
Solution Approach 1:
The patent creates simplified representations (copies) of user preferences at the token instance level. Instead of implementing complex real-time analysis of all content elements, the system builds preference models based on measured affective responses to token instances, which can then be used to guide personalized content delivery without requiring ongoing complex processing, thus resolving the contradiction between adaptability and device complexity.
Data Source
AI summary
Described herein are embodiments of systems, method, and computer programs for estimating affective response to a token instance of interest utilizing a predicted affective response to a background token instance. The affective response is computed based on the difference between a measured affective response of the user and the predicted affective response. Where the measured affective response is a measurement of the affective response of the user to a simultaneous exposure to both the background token instance and the token instance of interest.