Media Preference Profiles to Correct Runaway Recommendations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing media recommendation systems are 'black boxes' that fail to provide users with access to derived media preferences, leading to a runaway recommendation problem where users are no longer recommended media aligned with their initial preferences, resulting in dissatisfaction and excessive resource consumption.
Innovation Solution
A media presentation management system that processes express and behavior information to generate a media presentation profile, allowing users to observe and update their preferences in real-time, thereby aligning recommendations with current preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a media recommendation system operates as a black box without providing access to derived preferences, then the system complexity is reduced and ease of operation is improved, but the reliability of recommendations deteriorates due to the runaway recommendation problem
Solution Approach 1:
The patent implements feedback by displaying derived media preferences to users and allowing them to provide feedback through preference modifications. This feedback loop enables the system to correct runaway recommendations by comparing derived preferences with user confirmations or corrections, thereby maintaining recommendation reliability while keeping the system easy to operate.
Solution Approach 2:
The patent introduces an intermediary layer (the displayed derived preferences) between the black box recommendation algorithm and the user. This intermediary allows users to observe and validate the reasoning behind recommendations without exposing the complex algorithm, maintaining ease of operation while improving reliability through user awareness and feedback.
2Adaptability or versatility
If a media recommendation system continuously processes user behavior to generate recommendations, then the adaptability of recommendations is improved, but the resource consumption on the user device increases
Solution Approach 1:
The patent extracts and displays only the essential derived preferences to the user rather than processing and presenting all raw behavior data. This extraction approach maintains adaptability by focusing on key preference indicators while reducing the computational burden and energy consumption on the user device.
Solution Approach 2:
The patent applies partial action by processing and displaying only the most relevant derived preferences rather than continuously analyzing all user behaviors in real-time. This selective processing maintains adequate adaptability for recommendation quality while significantly reducing the energy consumption required for continuous full-scale analysis.
3Loss of information
If a media recommendation system provides detailed derived preference information to users, then the loss of information is reduced and user understanding is improved, but the device complexity increases
Solution Approach 1:
The patent creates a simplified copy or representation of the complex derived preferences that can be displayed to users without exposing the underlying algorithmic complexity. This copying approach reduces information loss by presenting essential preference data in an accessible format while maintaining low device complexity through the use of streamlined display structures.
Data Source
AI summary
In some implementations, a system may determine based on express preference information associated with media presentation on a user device, and based on behavior information associated with at least first media presented on the user device, derived preference information associated with media presentation on the user device. The system may generate a media presentation profile that indicates at least some of the express preference information and at least some of the derived preference information. The system may send, to the user device, the media presentation profile for presentation on the user device in association with the first media.