Content Selection Criteria via Entity Co-occurrence Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Content providers face challenges in selecting relevant additional content for web pages, as existing methods lack efficiency in suggesting criteria that enhance coverage and engagement.
Innovation Solution
A system and method that automatically suggest additional content selection criteria by identifying co-occurring entities across multiple content providers, evaluating their performance metrics, and associating them with the content of the first provider if they meet certain thresholds, such as click-through rates or frequency of occurrence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If content providers manually select content selection criteria, then they can control content quality, but the coverage and efficiency of content selection is limited
Solution Approach 1:
The system automatically generates content selection criteria by analyzing entity co-occurrence patterns across multiple content providers. The aggregator identifies entities that frequently appear together in content groups, and the selection criteria generator automatically creates optimized criteria without requiring manual intervention from content providers, thereby improving efficiency while maintaining manageable system complexity through automated processes.
Solution Approach 2:
The system performs preliminary analysis of entity co-occurrence patterns across content groups before content selection is needed. By pre-identifying frequently co-occurring entities and generating selection criteria in advance, the system prepares optimized content selection parameters that can be quickly applied when content needs to be selected, improving overall productivity without adding operational complexity.
2Adaptability or versatility
If content providers use limited content selection criteria, then the system remains simple, but coverage and engagement are reduced
Solution Approach 1:
The system analyzes entity co-occurrence patterns across multiple content providers and content groups to identify universally applicable selection criteria. By finding entities that frequently appear together across different content groups, the system generates versatile selection criteria that can be applied broadly to improve coverage while maintaining reasonable complexity through pattern-based generalization rather than provider-specific customizations.
Solution Approach 2:
The system dynamically adjusts content selection criteria based on entity co-occurrence frequency and performance metrics. By changing the parameters of selection criteria according to measured performance and observed entity patterns, the system adapts to improve coverage and engagement while managing complexity through data-driven parameter optimization rather than structural complexity.
3Measurement precision
If the system analyzes co-occurrence across multiple content providers, then it can identify better content patterns, but the processing time and computational resources increase
Solution Approach 1:
The system performs preliminary analysis of entity co-occurrence patterns across content groups to pre-identify frequently appearing entity combinations. By conducting this analysis in advance and storing the results, the system can quickly generate content selection criteria without performing full multi-provider analysis each time content needs to be selected, thereby improving measurement precision while reducing the time loss for criteria generation.
Solution Approach 2:
The system identifies and copies successful entity co-occurrence patterns from multiple content providers to generate content selection criteria. By replicating proven patterns across providers rather than analyzing each provider individually in real-time, the system achieves high measurement precision through pattern recognition while minimizing processing time through efficient pattern copying and application.
Data Source
AI summary
The present disclosure relates to systems and methods for refining content selection criteria to facilitate content selection via a computer network. In some embodiments, a tool identifies a first entity used to select content of a first content provider. The first entity can correspond to a canonical stable identifier in a database. The tool can identify a plurality of sets of entities that each comprise the first entity and an additional entity different from the first entity. Each of the plurality of sets of entities can be associated with content of a corresponding content provider. The tool can determine a performance metric for the additional entity. The tool can compare the performance metric with a performance threshold. The tool can associate the additional entity with the content of the first content provider based on the comparison.