Query Rewriting Using Session Context for Search Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search systems often fail to accurately return results that align with a user's intended query, as they do not effectively utilize previous search queries within a session to improve query relevance.
Innovation Solution
The system generates candidate query rewrites by combining the current search query with prior search queries from the same session, scoring them based on factors like timestamp, popularity, and referential terms, and selects the best rewrite to provide more relevant search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system uses only the current search query without prior queries, then the system complexity is low, but the query relevance to user intent deteriorates
Solution Approach 1:
The system performs preliminary actions by obtaining and storing prior search queries from the user session before processing the current search query. These prior queries are retained and made available for generating candidate rewrites, allowing the system to leverage historical context without adding complex real-time processing during the main search operation.
Solution Approach 2:
The query processing is segmented into distinct components: obtaining prior queries, generating candidate rewrites by combining current and prior queries, scoring the candidates, and selecting the best rewrite. This segmentation allows the system to manage complexity by breaking down the overall task into manageable, modular steps.
2Measurement precision
If the system generates multiple candidate query rewrites using prior queries, then the query relevance improves, but the processing time increases
Solution Approach 1:
The system generates multiple candidate query rewrites (excessive action) by combining the current query with each prior query in the session. Rather than attempting to find the perfect rewrite through exhaustive analysis, the system generates several candidates and uses scoring to select the most relevant one, balancing thoroughness with efficiency.
Solution Approach 2:
The system changes parameters by weighting candidate rewrites based on the age of prior queries (using timestamps) and other factors. This parameter-based scoring allows the system to efficiently evaluate and rank multiple candidates without requiring equally detailed analysis of each one, reducing overall processing time.
3Measurement precision
If the system weights candidate rewrites based on prior query age, then the query relevance improves, but the computational complexity increases
Solution Approach 1:
The system applies parameter changes by introducing weighting factors based on prior query characteristics such as timestamp (age) and popularity. These parameters are integrated into the scoring function to rank candidate rewrites, allowing the system to capture contextual relevance without requiring complex semantic analysis of each candidate.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for natural language processing. One of the methods includes receiving a search query from a user during a user session; obtaining a plurality of prior search queries by the user received during the user session; generating a plurality of candidate query rewrites, wherein the candidate query rewrites are derived from the search query and the plurality of prior search queries by the user; scoring each candidate query rewrite, wherein scoring each candidate rewrite includes determining a quality of each candidate query rewrite based on an analysis of search results responsive to the candidate query rewrite; selecting a candidate query rewrite having a score that satisfies a threshold value; and providing search results responsive to the selected candidate query rewrite.