Iterative Query Expansion for Cross-Community Document Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing query expansion methods fail to effectively retrieve a full set of relevant documents across different communities using community-specific vocabulary, often resulting in irrelevant results or incomplete document sets due to limited iterative expansion strategies.
Innovation Solution
A system and method for iterative, corpus-driven query expansion that receives an initial search query, identifies enrichment terms from retrieved documents, and generates subsequent search queries until a stopping criterion is met, incorporating high-leverage query terms to retrieve a comprehensive set of documents regardless of community-specific vocabulary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If query expansion uses community-specific vocabulary, then documents from different communities can be retrieved, but relevant documents may be missed due to vocabulary differences
Solution Approach 1:
The system implements iterative query expansion where each iteration uses the results from the previous iteration to generate new search queries. The feedback loop continuously refines the search by identifying enrichment terms from retrieved documents and using them to expand the query, ensuring comprehensive coverage across different communities while maintaining retrieval reliability.
Solution Approach 2:
The system performs preliminary query expansion by identifying enrichment terms from the initial set of retrieved documents before conducting subsequent searches. This preliminary action prepares expanded query terms that capture community-specific vocabulary and concepts, enabling the system to retrieve relevant documents across different communities effectively.
2Device complexity
If iterative expansion is performed for a fixed number of iterations, then the search process is simple, but the stopping criterion may not be met optimally
Solution Approach 1:
The system uses feedback from each iteration to determine whether the stopping criterion is met. The feedback mechanism evaluates the results of each query expansion iteration and decides whether to continue or terminate the process, ensuring optimal stopping accuracy while maintaining process simplicity through automated decision-making.
Solution Approach 2:
The search process dynamically adjusts the number of iterations based on whether the stopping criterion is met in each iteration. The system transitions from a static fixed-iteration approach to a dynamic conditional approach where the process continues or terminates based on real-time evaluation of the stopping criterion, optimizing both accuracy and simplicity.
3Quantity of substance
If enrichment terms are added to expand the search query, then more relevant documents are retrieved, but the query becomes more complex
Solution Approach 1:
The system segments the query expansion process into distinct iterations, where each iteration adds a specific set of enrichment terms to the query. This segmentation allows the system to manage query complexity by building up terms incrementally rather than adding all terms at once, while still retrieving a comprehensive quantity of relevant documents.
Solution Approach 2:
The system performs preliminary identification and ranking of enrichment terms before adding them to the query. This preliminary action organizes the terms in a structured manner, allowing the system to expand the query with multiple relevant terms while maintaining manageable complexity through pre-organization and prioritization.
Data Source
AI summary
An embodiment for expanding a search query is provided. The embodiment may include receiving a stopping criterion for stopping a search. The embodiment may also include receiving an initial search query. The embodiment may further include submitting the initial search query to an information retrieval system. The embodiment may also include identifying enrichment terms from the retrieved initial set of documents. The embodiment may further include generating a subsequent search query that includes one or more enrichment terms from the retrieved initial set of documents. The embodiment may also include submitting the subsequent search query to the information retrieval system. The embodiment may further include determining whether the stopping criterion is met, and in response to determining the stopping criterion is not met, iterating identifying, generating, submitting steps until the stopping criterion is met.