Content Recommendation System Using Segmented Viewing Duration Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing content recommendation systems for television and radio programs face challenges in accurately predicting user preferences due to the passive nature of television viewing, leading to suboptimal user preference models and reduced recommendation accuracy, especially when relying on implicit feedback which can be unreliable.
Innovation Solution
A recommendation apparatus and method that monitor the presentation of content items, determine preference values based on the duration watched relative to the total duration, and calculate confidence values for missed sections, generating a user preference model for improved recommendations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If implicit feedback is used to collect user preferences by monitoring user actions, then user input effort is reduced and ease of operation is improved, but the reliability of preference data deteriorates due to passive viewing behavior
Solution Approach 1:
The patent segments the content item into multiple sections (beginning section, middle section, end section) and monitors user viewing behavior in each segment separately. By analyzing which sections are viewed and in what order, the system can distinguish between genuine preference and passive viewing patterns, thereby improving reliability while maintaining ease of operation.
Solution Approach 2:
The system dynamically adjusts the weighting and interpretation of viewing data based on the temporal position within the content item. Viewing behavior in different sections (beginning vs. end) is treated differently to capture user intent more accurately, resolving the contradiction between passive viewing and reliable preference detection.
2Device complexity
If user preferences are collected by monitoring viewing duration, then system complexity is reduced, but measurement precision of user preferences deteriorates
Solution Approach 1:
The viewing duration is segmented into distinct temporal intervals (beginning section, middle section, end section) rather than treating it as a single continuous metric. This segmentation allows the system to capture more nuanced preference information about which parts of the content are viewed, improving measurement precision without significantly increasing system complexity.
Solution Approach 2:
The patent adds a temporal dimension to the viewing duration measurement by analyzing the sequence and duration of section viewing. Instead of a single duration metric, the system measures duration across multiple temporal segments, thereby improving precision while keeping the implementation relatively simple.
3Productivity
If only partial content items are monitored, then productivity of preference collection is improved, but loss of information about complete viewing behavior increases
Solution Approach 1:
The content item is divided into manageable sections (beginning, middle, end) that can be monitored independently. This segmentation allows the system to efficiently collect preference data from partial viewing sessions without losing critical information, as each section provides meaningful insight into user preference regardless of whether the entire content is viewed.
Data Source
AI summary
A recommendation apparatus comprises a monitoring processor which monitors the presentation of content items. A sample processor determines preference data for different content items by performing the steps of determining a preference value for a content item presented by the presentation unit in response to a first duration for a first section of the content item being presented relative to a total duration of the content item, and if the first duration is less than the total duration, determining if a second section of the content item not being presented corresponds to at least one of an end section and a begin section of the content item; and if so determining a confidence value for the preference value in response to a second duration of the second section. The preference data is used as training data for determining a user preference model which is then used to generate recommendations.