Content Hierarchy Relevance Propagation for Accurate Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Web search technologies fail to fully utilize configurational information in complex content items, leading to inefficient and inaccurate search results, particularly in query-dependent processes, and there is a need for improved computational efficiency and distribution of processing burdens across systems.
Innovation Solution
The methods and apparatuses determine topical relevance of content items within hierarchies by considering interrelationships among content items, using a non-associative juxtaposition operator and external influences, with optimizations for parallel processing and pre-search-time data preparation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If configurational information is systematically applied in search processes, then search accuracy is improved, but computational resources are consumed
Solution Approach 1:
The patent applies preliminary action by performing pre-search-time processing to prepare configurational information, content hierarchies, and search expressions before actual search execution. This includes building content graphs, determining topical relevance, and pre-processing documents during idle periods, thereby reducing computational burden during search-time operations while maintaining high search accuracy
2Measurement precision
If complex search expressions are evaluated over complex content hierarchies, then more accurate and focused search results are achieved, but processing time increases
Solution Approach 1:
The system performs preliminary evaluation of complex search expressions and builds content graphs during pre-search-time processing, storing intermediate results and topical relevance relationships. During search-time, these pre-computed structures enable rapid evaluation of complex expressions without re-processing the entire content hierarchy, significantly reducing processing time while maintaining accuracy
Solution Approach 2:
The patent segments the search process into distinct phases: pre-search-time processing (building content graphs, determining topical relevance, pre-processing documents) and search-time processing (evaluating expressions, retrieving results). This segmentation allows computationally intensive operations to be performed in advance, breaking down the overall processing into manageable segments that can be optimized independently
3Device complexity
If all search processing is performed at search-time, then system simplicity is maintained, but productivity decreases
Solution Approach 1:
The patent divides the search system into multiple components with distinct responsibilities: pre-search-time processing module (builds content graphs, determines topical relevance), search-time processing module (evaluates expressions, retrieves results), and distribution layer (coordinates processing across systems). This segmentation enables the system to achieve high productivity through parallel processing while maintaining manageable complexity through clear separation of concerns
Solution Approach 2:
The system implements preliminary action by performing document processing, content graph construction, and topical relevance determination during pre-search-time operations. This allows the search system to maintain simplicity during actual search operations while achieving high productivity through pre-computed structures and distributed processing of preparatory tasks
4Device complexity
If computational burden is concentrated in a single system, then system simplicity is maintained, but scalability is limited
Solution Approach 1:
The patent implements a universal search processing framework where multiple systems can perform identical pre-search-time and search-time operations. Each system in the distributed network maintains the same capabilities to build content graphs, evaluate expressions, and process documents, enabling seamless scaling from single-system to multi-system deployments without increasing architectural complexity
Solution Approach 2:
The system segments computational tasks into independent units that can be distributed across multiple systems: document processing, content graph construction, topical relevance determination, and search expression evaluation. Each system handles its own segment of the workload while maintaining consistency through standardized data formats and protocols, enabling linear scalability
Data Source
AI summary
Method, storage medium and apparatus associated with content preparation and/or selection are disclosed herewith. In embodiments, an apparatus may include a content preparation or selection module to receive, as part of a content preparation or selection, a representation of a concept; and determine, for the content preparation or selection, topical relevance of a content item (N) of a content hierarchy to the concept. The content hierarchy may include a plurality of content items, including N, and determining the topical relevance of N to the concept includes determination of propagation of influence among content items of the content hierarchy subject to constraints that bound the number of operations performed by the computing system to determine the propagation of influence among content items of the content hierarchy with a chosen constant. Other embodiments may be disclosed and/or claimed.