Personalized Content Channel Recommendation Engine
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing personalized content channels lack surprising recommendations, making them predictable for users, and existing recommender systems require significant computation.
Innovation Solution
A method and system using a filter with multiple selection criteria to generate recommendations for content items that partially satisfy these criteria, allowing for surprising yet user-preference-matching suggestions, with a recommender engine that adjusts criteria based on user feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a filter with strict multiple selection criteria is used to provide personalized content channels, then content accuracy and user preference matching are improved, but the content becomes predictable and lacks surprising recommendations
Solution Approach 1:
The patent applies partial action by allowing content items to partially satisfy filter criteria rather than requiring full satisfaction. The recommender engine identifies content that meets some but not all selection criteria, generating surprising recommendations that still align with user preferences. This resolves the contradiction by accepting partial matches instead of demanding complete criterion satisfaction.
2Adaptability or versatility
If complex recommender systems are used to generate surprising recommendations, then recommendation diversity is improved, but computational requirements increase significantly
Solution Approach 1:
The patent introduces an intermediary approach by using a recommender engine that acts as a mediator between strict filters and content recommendations. Instead of directly computing complex recommendations, the system uses the recommender engine to identify content that partially satisfies filter criteria, reducing computational overhead while maintaining recommendation diversity.
3Ease of operation
If users must manually browse recorded programs to select items for deletion, then storage management control is improved, but user time and effort increase
Solution Approach 1:
The patent applies self-service by implementing automatic deletion strategies that manage storage without requiring manual user intervention. The system automatically identifies and deletes content based on predefined criteria such as watch completion status or episode recency, freeing users from manual browsing while maintaining storage management control.
Data Source
AI summary
Method and system (100) for generating a recommendation for at least one further content item is disclosed. A personalized content channel is enabled to play out a plurality of content items (programs) complying with multiple selection criteria. At least one further content item is recommended by a recommender engine (107), the at least one further content item complying with fewer of the multiple criteria. In an embodiment, at least one of the at least one recommended further content item is selected and the multiple selection criteria are adjusted by a scheduler (109) on the basis of at least one characteristic of the selected recommended further content item.