Broadcast Content Ranking With Item-Level Preference Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to accurately estimate user preferences for broadcast content beyond genre-level analysis, as viewing history often misrepresents individual content preferences.
Innovation Solution
An information processing device that assigns ranking and preference scores to broadcast content based on viewing history and user preferences, using AI models to tag content and combine scores for optimized presentation order.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If user preferences are estimated from viewing history, then individual content preferences can be identified, but viewing history may misrepresent actual preferences leading to inaccurate recommendations
Solution Approach 1:
The patent segments the preference estimation process into multiple independent components: viewing history analysis, genre-level preference extraction, and item-level preference scoring. By dividing the estimation task into separate analytical stages, the system can identify discrepancies between viewing patterns and actual preferences, thereby improving measurement precision while accounting for viewing history limitations
Solution Approach 2:
The patent introduces genre-level preferences as an intermediary layer between viewing history and individual content preferences. This intermediary abstraction helps bridge the gap by first establishing broader preference patterns from viewing history, then refining to specific content preferences, thereby improving reliability while maintaining accuracy
2Productivity
If broadcast content is arranged by popularity ranking, then widely viewed content is presented, but content aligned with individual user preferences may be lost
Solution Approach 1:
The patent implements a dynamic content arrangement system that adapts the presentation order based on user preferences. Instead of a fixed popularity-based ranking, the system dynamically adjusts content ordering by combining popularity scores with user-specific preference scores, thereby maintaining presentation efficiency while improving adaptability to individual users
Solution Approach 2:
The patent applies different weighting strategies to different content items based on user preferences. High-preference content receives greater weight in the ordering calculation, while low-preference content receives less weight. This local quality adjustment ensures that content presentation is optimized for each user's specific preferences rather than applying a uniform popularity-based approach
Data Source
AI summary
An information processing device includes: an obtainer that obtains a ranking score assigned to each of a plurality of items of target broadcast content; a preference outputter that outputs preference information based on a viewing history of one or more items of broadcast content viewed by a user, the preference information being related to a preference of the user for each of the one or more items of broadcast content; an evaluator that assigns a preference score to each of the plurality of items of target broadcast content; and an order outputter that outputs order information based on the ranking score and the preference score, the order information being related to an order of the plurality of items of target broadcast content arranged according to the preference of the user.