Multi-Term Contextual Tag Propagation for Digital Content Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital visual systems fail to efficiently and accurately search for and retrieve contextually relevant digital content due to inadequate utilization of tags, which lack semantic completeness and context, leading to irrelevant search results and inefficient resource usage.
Innovation Solution
A digital content contextual tagging system that utilizes a multi-modal learning framework to mine relevant tag combinations from user behavior data and propagate multi-term contextual tags across an image database, ensuring accurate and efficient searching by determining and associating tags based on user search queries and image similarities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional digital visual systems utilize independent tags for searching, then the system can process search queries, but the search results lack contextual accuracy and relevance
Solution Approach 1:
The patent combines multiple independent tags into a single multi-term contextual tag that preserves the semantic relationships between terms. Instead of treating tags as independent keywords, the system merges them into contextual units that maintain the original context, thereby improving search accuracy without losing contextual information.
Solution Approach 2:
The system creates composite tags by combining multiple simple tags into a structured multi-term tag. This composite approach allows the tag to carry both the individual term meanings and their contextual relationships, enabling more precise search matching while preserving the full contextual information.
2Productivity
If conventional systems conduct additional search queries to compensate for contextual inaccuracies, then more results can be retrieved, but computational resources are inefficiently utilized
Solution Approach 1:
The system performs preliminary tagging by creating accurate multi-term contextual tags during the content ingestion phase. This preliminary action ensures that when search queries are executed, the system can directly match against contextually accurate tags without needing to conduct additional compensatory search queries, thereby reducing computational resource consumption.
Solution Approach 2:
The system uses user interaction feedback to refine and update multi-term contextual tags. By incorporating feedback from actual search behavior and user selections, the system continuously improves tag accuracy, which enhances retrieval effectiveness while reducing the need for resource-intensive additional queries.
3Measurement precision
If human annotators are used to tag digital content accurately, then contextual relevance is maintained, but the cost and time consumption increase significantly
Solution Approach 1:
The system enables automatic generation of multi-term contextual tags through machine learning algorithms that analyze content and generate contextually accurate tags without human intervention. This self-service approach maintains high tagging accuracy while eliminating the time consumption and costs associated with human annotators.
Solution Approach 2:
The system introduces an intermediary machine learning model that acts as a bridge between raw content and final tags. This intermediary automatically processes content to generate multi-term contextual tags with high accuracy, replacing the need for human annotators while maintaining quality standards.
4Ease of operation
If conventional systems utilize bag-of-words models for tag association, then tag processing is simplified, but contextual relationships between tags are lost
Solution Approach 1:
The system segments the tagging process into two distinct stages: first, generating multi-term contextual tags that preserve relationships, and second, processing these tags for search operations. This segmentation allows the system to maintain contextual information in the tag structure while using efficient processing methods for retrieval operations.
Data Source
AI summary
Systems, methods, and non-transitory computer-readable media are disclosed for determining multi-term contextual tags for digital content and propagating the multi-term contextual tags to additional digital content. For instance, the disclosed systems can utilize search query supervision to determine and associate multi-term contextual tags (e.g., tags that represent a specific concept based on the order of the terms in the tag) with digital content. Furthermore, the disclosed systems can propagate the multi-term contextual tags determined for the digital content to additional digital content based on similarities between the digital content and additional digital content (e.g., utilizing clustering techniques). Additionally, the disclosed systems can provide digital content as search results based on the associated multi-term contextual tags.