Video Ad Selection Using Time-Weighted Viewing Intervals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing content recommendation systems fail to effectively utilize consumption time intervals and similarity between previously consumed content to enhance recommendation accuracy.
Innovation Solution
Adjust contribution factors for previously consumed content based on time differences relative to consumption intervals, such as multiples of 24 hours and 168 hours, to determine recommendation scores, leveraging viewing history data to improve content suggestions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If contribution factors for previously consumed content are adjusted based on time differences relative to consumption intervals, then recommendation accuracy is improved, but system complexity increases
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting contribution factors based on time differences between content consumption events. The system modifies the weight of previously consumed content in recommendation calculations according to whether the time difference aligns with observed consumption intervals (e.g., 24 hours, 168 hours), thereby improving recommendation accuracy without fundamentally changing the system architecture
Solution Approach 2:
The system implements dynamics by making contribution factors adaptive rather than static. The contribution factor for each previously consumed content item is dynamically adjusted based on the time difference to the current recommendation moment, allowing the system to respond to changing user behavior patterns while maintaining a relatively simple underlying recommendation engine
2Productivity
If viewing history data is leveraged with time-based weighting, then user engagement is improved, but data processing requirements increase
Solution Approach 1:
The patent applies local quality by differentiating the treatment of individual content items in the viewing history based on their temporal characteristics. Each content item receives a localized adjustment to its contribution factor according to its specific time difference from the current moment, allowing important recent items to have higher weight while older items naturally decay, thereby improving user engagement with computationally efficient targeted processing
Data Source
AI summary
Aspects described herein describe selecting advertisements for insertion into video content. An advertisement may be selected based on when content was previously consumed.