Multi-Shot Scheduling System for Media Playlist Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current media broadcasting technologies schedule content items individually, leading to suboptimal selection decisions for later content items due to earlier selections, resulting in less than ideal results.
Innovation Solution
The Multi-Shot Scheduling System uses processing circuitry to generate and select candidate playlists by assigning content items to positions based on score and index values, with a demand-based goal-driven algorithm, and restricts position index values to improve selection efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If content items are scheduled individually, then the scheduling process is simple, but the selection decisions for later content items are negatively impacted by earlier selections
Solution Approach 1:
The patent segments the scheduling process into multiple independent shots, where each shot selects content items for a specific time period without being constrained by previous selections. This segmentation allows each shot to make optimal selection decisions independently, resolving the contradiction between simple individual scheduling and high-quality selection decisions.
2Reliability
If multiple candidate playlists are generated and evaluated, then the quality of selected playlists improves, but the processing time increases
Solution Approach 1:
The patent performs preliminary actions by pre-generating multiple candidate playlists with different content item assignments before final selection. By preparing multiple options in advance and evaluating them systematically, the system ensures high-quality playlist selection while managing processing time through efficient pre-computation.
Solution Approach 2:
The patent changes parameters by generating candidate playlists with varying content item assignments and evaluating them based on aggregate scores. By systematically varying the parameters of content item selection across multiple candidates and selecting based on optimized criteria, the system achieves high-quality results while controlling processing time through parameter-based evaluation.
3Productivity
If position index values are restricted to a predetermined range, then the number of candidate playlists is reduced, but the flexibility of content assignment is limited
Solution Approach 1:
The patent applies local quality by restricting position index values to predetermined ranges for specific positions within candidate playlists. This localized constraint approach improves selection efficiency for certain positions while maintaining flexibility for others, resolving the contradiction between productivity and adaptability by applying different quality levels to different parts of the assignment process.
Data Source
AI summary
A method for use in an automated media scheduling system includes generating a shots representing potential playlists of media items. Each of the shots includes multiple shot positions. A shot-position-dependent score of a current content item being considered for insertion into a particular shot position of a particular shot is generated, and indicates how closely the current content item conforms to a predetermined set of factors associated with the particular shot position of the particular shot. The current content item is inserted into the particular shot position of the particular shot based, at least in part on a comparison of the shot-position-dependent score of the current content item with shot-position-dependent scores of other content items, and the shot positions of each of the shots are populated with eligible content items. One of the shots is selected for scheduling based on aggregate scores of eligible content associated with each individual shot.