Semantic Query Expansion via Decaying Radius Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search engines provide limited assistance to users formulating queries, especially when searching unfamiliar topics, as they do not effectively suggest semantically related queries to aid in query formulation and result selection.
Innovation Solution
A computing system expands an initial set of queries using a breadth-first search, generating semantically similar terms and caching them to enhance search result selection and advertisement display, by iteratively selecting related terms and applying a decaying radius to manage the number of suggestions until a threshold is reached.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional search engines filter related queries based on user context data, then the system complexity is reduced, but the quality of query assistance deteriorates for unfamiliar topics
Solution Approach 1:
The system pre-generates and caches an expanded set of semantically similar queries for common search topics before users actually search. This preliminary expansion creates a ready-to-use knowledge base that assists users with unfamiliar topics without requiring complex real-time analysis of their context data.
Solution Approach 2:
The patent introduces an intermediary expanded query set that acts as a bridge between the user's initial query and the final search results. This intermediary layer provides semantic alternatives and related terms that guide users in formulating better queries without exposing the underlying system complexity.
2Ease of operation
If the expansion process includes more semantically similar terms, then the query assistance quality improves, but the data processing volume increases
Solution Approach 1:
The system applies semantic expansion selectively to specific query verticals and topics that benefit most from assistance. Rather than uniformly expanding all queries, the system identifies high-value areas where users commonly struggle with unfamiliar terminology and applies expansion there, maintaining quality while controlling data volume.
Solution Approach 2:
The expanded query sets are pre-computed and cached during off-peak times, transforming the data processing from a real-time burden into a manageable batch operation. This allows the system to generate comprehensive semantic expansions without impacting real-time performance.
3Loss of information
If the breadth-first search expands queries iteratively, then the semantic coverage improves, but the computation time increases
Solution Approach 1:
The system performs the computationally intensive breadth-first search iteratively in advance, building comprehensive semantic expansion maps before users need them. The pre-computed expanded sets capture multi-level semantic relationships that would be expensive to generate on-demand.
Solution Approach 2:
The semantic expansion process operates periodically or on-demand for new verticals, rather than continuously for every query. This batch processing approach allows thorough semantic exploration without occupying system resources during normal operation.
4Adaptability or versatility
If the system provides expanded query sets to researchers, then the research capability improves, but the system resource consumption increases
Solution Approach 1:
The system enables researchers to independently access and utilize pre-computed expanded query sets for their own analysis and experimentation. This self-service capability allows research activities to proceed using existing system resources without requiring additional real-time computational support.
Data Source
AI summary
Query suggestions are generated based on a Breadth-First-Search having a configurable decaying radius. A computer system receives an initial set of semantically similar queries. The computer system expands the set to include related terms. The set of related terms is included in the initial set. The expansion process is repeated for each query or related term in the set. The radius may be reduced for each subsequent related term added to the query. The process may stop when the radius reaches a specified threshold, e.g., a predetermined number of queries or terms for the set is reached. The set includes the related terms and search queries. The set may be used for, among other things, suggesting related terms to a researcher, improving search engine performance, or selecting appropriate advertisements.