Query Synonym Generation via Online Decompounding and Transitivity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search technologies fail to effectively identify and utilize context-specific synonyms for search query terms, leading to suboptimal search results due to reliance on pre-constructed synonym lists and contextual dependency, which are often language-restricted and contextually inaccurate.
Innovation Solution
A method that analyzes user query logs and contextual data to identify and rank context-specific synonyms by evaluating query pairs, frequency of substitution, and shared search results, allowing for dynamic alteration of search queries to improve relevance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If pre-constructed synonym information from thesauruses or lexical databases is used, then synonym identification is simplified, but the system becomes expensive to construct and restricted to one language
Solution Approach 1:
The system automatically constructs synonym information by analyzing query logs and document data itself, rather than relying on pre-constructed thesauruses. The synonym identification module extracts synonym relationships from actual usage patterns in search queries and documents, making the system self-sufficient and language-agnostic
Solution Approach 2:
The system pre-processes query logs and document data to build synonym information before actual search operations. By analyzing historical query data and document correlations in advance, the system prepares synonym relationships that can be quickly retrieved during search operations
2Quantity of substance
If conventional synonym lists from thesauruses are used, then synonym availability is improved, but contextual accuracy deteriorates because synonyms are not context-specific
Solution Approach 1:
The system makes synonym quality context-dependent by analyzing the specific query context, document domain, and usage patterns. Instead of applying uniform synonym relationships, the system adapts synonym selection to local contextual factors such as query type, document category, and co-occurring terms
Solution Approach 2:
The synonym information is dynamically constructed and updated based on actual query logs and document data rather than being static. The system continuously learns from new query patterns and document additions, making synonym relationships adaptive and context-aware
3Adaptability or versatility
If related words are clustered together, then word association is improved, but synonym accuracy deteriorates because related words are not necessarily synonyms
Solution Approach 1:
The system uses feedback from search results and query refinement patterns to distinguish true synonyms from merely related words. By analyzing whether substituting one word for another maintains search result quality and user query intent, the system learns to differentiate between association and synonymy
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for query synonym expansion. One method includes receiving a query including a first compound term, and in response to receiving the query, performing the following operations before search results responsive to the query are identified: generating one or more splits of the first compound term, wherein each split divides the compound term into two or more subterms, assigning a score to each subterm of each split, determining an overall score for each split from the scores for the subterms of the split, selecting one or more of the one or more splits according to the overall score for each split, and augmenting the query with the subterms of each selected split.