Query Expansion via Session Vote Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional query expansion methods lead to excessive searching times and increased system resource occupancy due to repeated searches and reduced system response speed, as they often require multiple matches between query and index expansion terms.
Innovation Solution
A query expansion method and system that normalizes queries using session information from search logs, determining a normalized query by calculating vote and click similarity degrees to reduce unnecessary searches and improve response speed and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional query expansion methods are used to expand queries at query end and index end, then search result accuracy is improved, but search times and system resource occupancy increase excessively
Solution Approach 1:
The patent performs preliminary normalization of queries by determining a normalized query from the original query before executing the search. This normalized query is then used for expansion matching against index expansion terms. By preparing and normalizing the query in advance, the system avoids repeated search operations and reduces the time required to match queries with expansion terms, while still maintaining accurate search results through the normalization process
2Adaptability or versatility
If multiple expansion terms are selected and matched sequentially, then comprehensive search coverage is achieved, but system response speed decreases due to repeated searching
Solution Approach 1:
The system determines a normalized query from the original query before initiating the search process. This normalized query serves as a standardized representation that can be efficiently matched against index expansion terms. By preparing the normalized query in advance, the system achieves comprehensive search coverage through the normalization process while reducing the time required for subsequent matching operations, thereby improving system response speed
3Loss of information
If queries are expanded at both query end and index end simultaneously, then query information is enriched, but system resource occupancy increases
Solution Approach 1:
The patent determines a normalized query from the original query before executing the search operation. This normalized query is then used for expansion matching against index expansion terms. By performing the normalization and expansion in a streamlined manner with the normalized query, the system enriches query information effectively while reducing the computational resources required compared to traditional simultaneous expansion approaches
Data Source
AI summary
The present disclosure provides example query expansion methods and systems. A query input by a user is received. A normalized query of the query is determined according to the query. The normalized query is used as an expansion term of the query to implement query expansion. For example, session information in a search log of the user is obtained. All queries appearing in a single session are obtained and a number of votes for each query is calculated or counted. A vote similarity between a single query and a target query is determined and a correlation degree between the single query and the target query is determined according to the vote similarity. A normalized query of the target query is determined according to the correlation degree. The present techniques accurately expand queries and reduce query times, thereby improving system response speed and processing efficiency.