Visual Tag Pattern Detection via Taxonomy Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems fail to effectively identify emerging visual media viewing patterns, which are crucial for tailoring content and advertising to users, as they rely on analyzing all visual tags individually rather than filtering specific tags and their combinations.
Innovation Solution
A method and system that utilize a taxonomy-based filter to select specific visual tags and generate pattern candidates from these tags, evaluating consumption metrics to rank emerging viewing patterns, allowing for more accurate identification of trends.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all visual tags are analyzed individually, then comprehensive coverage of visual media is achieved, but the system complexity and computational burden increase significantly
Solution Approach 1:
The patent segments the visual tag analysis process into distinct stages: initial tag generation, taxonomy-based filtering, pattern candidate generation, and metric evaluation. This segmentation allows the system to process visual tags in manageable stages rather than analyzing all tags simultaneously, reducing system complexity while maintaining comprehensive coverage.
Solution Approach 2:
The patent extracts and removes irrelevant visual tags from the complete set using taxonomy-based filtering. By taking out only the necessary tags and their combinations for pattern analysis, the system reduces computational burden and complexity while focusing on the most relevant visual patterns for emerging trend identification.
2Measurement precision
If all visual tags and their combinations are evaluated, then complete pattern detection is achieved, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary filtering of visual tags using taxonomy-based criteria before generating pattern candidates. This preliminary action eliminates irrelevant tags early in the process, so that subsequent pattern evaluation focuses only on promising candidates, significantly reducing processing time while maintaining detection completeness.
Solution Approach 2:
The patent evaluates a selective subset of pattern candidates rather than all possible tag combinations. By applying partial action to the most relevant patterns identified through filtering and ranking, the system achieves effective pattern detection without the excessive computational cost of exhaustive analysis.
3Productivity
If taxonomy-based filtering is applied to select specific visual tags, then processing efficiency improves, but the risk of missing emerging patterns increases
Solution Approach 1:
The patent incorporates feedback mechanisms where consumption metrics are evaluated for filtered visual tags and pattern candidates. This feedback loop allows the system to verify whether filtering has eliminated important patterns, and to adjust the filtering criteria accordingly, thereby maintaining reliability while preserving processing efficiency.
Solution Approach 2:
The patent applies taxonomy-based filtering with multiple functions: it organizes visual tags hierarchically, filters relevant tags for efficiency, and simultaneously generates pattern candidates from the filtered set. This multi-functional approach ensures that filtering serves multiple purposes without compromising pattern identification reliability.
Data Source
AI summary
Systems, devices, media, and methods are presented for identifying emerging viewing patterns for visual media such as still images and videos. Emerging viewing patterns are identified by identifying visual tags for visual media viewed by users, selecting a subset of the tags by applying a taxonomy-based filter, generating pattern candidates from the subset, evaluating consumption metrics for each of the generated patterns, and ranking the generated pattern candidates responsive to the consumption metrics to identify emerging viewing patterns for the users.