Tag-Based Media Recommendation Interface for User Agency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing media recommendation systems fail to effectively incorporate user agency and serendipity, often relying on explicit tags or related tags, leading to a cumbersome process that transforms into a search engine rather than a recommendation engine, and do not adequately utilize user profiles for personalized recommendations.
Innovation Solution
A client device and recommender system that displays a recommendations list alongside a tag list, where each tag is associated with representative items relevant to the user, allowing users to interactively adjust the recommendations by adding or excluding tags, and using these interactions to provide more relevant media suggestions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If prior art methods search for explicit tags or related tags to provide recommendations, then user agency is incorporated, but the system becomes more like a search engine requiring many user actions
Solution Approach 1:
The patent introduces a dual-list interface with candidate items and candidate tags as intermediaries. Instead of directly searching through tags (which makes it feel like a search engine), the system presents both items and tags as selectable candidates. The user can select from either list, and the system processes these selections through a recommendation algorithm that considers the relationship between items and tags. This intermediary presentation layer transforms the direct tag-searching process into a more recommendation-oriented experience where users have agency but don't need to perform many sequential actions.
2Measurement precision
If prior art methods rely on tag accuracy for recommendations, then recommendation precision is maintained, but the system loses serendipity and user engagement
Solution Approach 1:
The patent implements dynamics by allowing the recommendation results to change based on user selections from the dual-list interface. The system doesn't rigidly follow pre-defined tag relationships but adapts in real-time as users select items or tags. This dynamic adjustment enables serendipitous discoveries while maintaining precision through the underlying tag-item relationships. The system can pivot between precision-driven and serendipity-driven recommendations based on user interaction patterns.
Solution Approach 2:
The system performs preliminary actions by pre-processing and organizing the relationship between items and tags before user interaction. Candidate items and candidate tags are prepared in advance based on existing data, but the final recommendation is determined through user selection. This preliminary organization maintains precision while the user's selective input introduces serendipity and personalization.
3Device complexity
If non-interactive browsing mode is used to display recommendations, then system complexity is reduced, but user engagement and personalization are limited
Solution Approach 1:
The patent segments the recommendation interface into distinct functional components: a candidate items list, a candidate tags list, and separate selection mechanisms for each. This segmentation allows the system to maintain relatively simple architecture while enabling rich user interaction. Each segment can be processed and updated independently, reducing overall system complexity while providing engaging user experience through selective interaction with items and tags.
Data Source
AI summary
A tag-based, user-directed media recommendation scheme is described herein. For example described herein is a recommender system (and method implemented by the recommender system) comprising: a recommendation engine configured to generate a recommendations list which includes a current set of recommendations for a user of a client device; and, a tag engine configured to: (1) receive the recommendations list; (2) obtain information about representative items associated with the user of the client device; (3) correlate the recommendations with tags; (4) correlate the representative items with the tags, (5) sort the tags into a tag list; and (6) provide the recommendations list and the tag list to the client device. The representative items include at least one of following: (1) an item previously purchased by the user; (2) an item previously watched by the user; (3) an item previously placed on a wish-list by the user; and (4) an item recommended by the recommendation engine for the user. Further described herein, are the client device and method implemented by the client device.