Query Pattern Graph Generation for Search Intent Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search systems often return tangentially relevant search results due to misinterpretation of user intent, leading to inefficiencies in information retrieval and increased user burden in finding desired information.
Innovation Solution
A system and method for generating query pattern graphs by analyzing query-document pairs, identifying query patterns, and assigning classifiers to associate user intent with web documents, thereby aggregating similar queries into a pattern graph that filters out irrelevant results and provides more accurate search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If search systems return all matching documents, then the quantity of search results increases, but the relevance and precision of search outcomes deteriorates
Solution Approach 1:
The patent segments the search result set by creating multiple query patterns from a single user query. Each query pattern represents a different interpretation or aspect of the user's information need. The system then retrieves documents for each pattern separately and combines them, effectively segmenting the search process to improve both quantity and relevance of results.
Solution Approach 2:
The patent adds a new dimension to search by generating query patterns that explore different semantic interpretations of the original query. Instead of returning documents based on a single query vector, the system creates multiple query vectors representing different dimensions of user intent, thereby expanding the search space while maintaining precision through targeted pattern matching.
2Loss of information
If users perform multiple search queries to find desired information, then the completeness of information retrieval improves, but the time and effort required increases
Solution Approach 1:
The patent performs preliminary action by automatically generating multiple query patterns before the user needs to execute them. The system anticipates different information needs and formulates corresponding queries in advance, then executes all patterns and combines results, saving the user the time and effort of manually performing multiple search iterations.
Solution Approach 2:
The system provides self-service by automatically generating and executing multiple query patterns without requiring user intervention. The patent implements an automated query expansion mechanism that independently creates alternative queries, retrieves documents, and synthesizes results, allowing the search system to serve itself rather than relying on the user to reformulate queries.
3Adaptability or versatility
If search systems interpret user intent broadly, then the coverage of search results increases, but the precision of matching user needs deteriorates
Solution Approach 1:
The patent applies local quality by creating query patterns with different levels of specificity tailored to local aspects of user intent. Each generated query pattern focuses on a particular interpretation or aspect of the original query, allowing the system to maintain broad coverage while ensuring each pattern precisely targets a specific information need with high relevance.
Data Source
AI summary
One general aspect is described that includes a computer implemented method for generating a pattern graph. The method may include accessing data pertaining to a corpus of web documents. The data may include a plurality of query-document pairs. The method may also include identifying at least one query pattern in the plurality of query-document pairs and the query pattern may be associated with a portion of web documents in the corpus. The method may also include identifying a plurality of sub-phrases in the at least one query pattern, determining, in the corpus of web documents, a plurality of other query patterns that include at least one of the plurality of sub-phrases, and assigning an classifier to the at least one query pattern and each of the plurality of other query patterns that include at least one of the sub-phrases.