Query Rewriting via Click Log Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search engines face challenges in addressing the query-document vocabulary mismatch, where user search queries do not yield desired results due to inherent biases in human-created lists and limitations of replacement terms, leading to unhelpful search results.
Innovation Solution
The system analyzes click log data to identify related search queries and supplemental strings based on dwell time, number of search result links clicked, and similarity between web page titles or URLs, incorporating these into a list for input into a statistical machine translation model to generate query rewriting candidate terms that reduce the query-document vocabulary mismatch.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If human-created lists and human-annotated thesauri are used for query rewriting, then the replacement terms may seem appropriate to human reviewers, but the query-document vocabulary mismatch is not solved and search result relevance deteriorates
Solution Approach 1:
The system uses click log data from user search behavior to automatically generate replacement term lists without human annotation. The system self-learns from actual user clicking patterns what terms lead to relevant documents, eliminating the need for manual thesauri creation while improving search relevance through real-world usage data
Solution Approach 2:
The system incorporates feedback from user click behavior and dwell time on search results to continuously refine and update replacement term lists. By monitoring which rewritten queries lead to successful document retrieval, the system adapts and improves its query rewriting effectiveness over time
2Device complexity
If conventional query rewriting methods are used, then the process is simple, but the query-document vocabulary mismatch persists and search effectiveness is limited
Solution Approach 1:
The system pre-processes and analyzes click log data to build comprehensive replacement term lists before they are needed for query rewriting. By performing this analysis in advance and storing the results in accessible data structures, the system makes complex preprocessing work happen beforehand, keeping the actual query rewriting process simple and fast
Solution Approach 2:
The system introduces click log data as an intermediary layer between the user's query and the document search. This intermediary data source provides empirically-derived replacement terms that bridge the vocabulary mismatch, allowing the system to maintain simple rewriting logic while achieving improved search effectiveness through data-mediated term substitution
Data Source
AI summary
Systems, methods, and computer media for identifying related strings for search query rewriting are provided. Session data for a user search query session in an accessed click log data is identified. It is determined whether a first additional search query in the session data is related to a first user search query based on at least one of: dwell time; a number of search result links clicked on; and similarity between web page titles or uniform resource locators (URLs). When related, the first additional search query is incorporated into a list of strings related to the first user search query. One or more supplemental strings that are related to the first user search query are also identified. The identified supplemental strings are also included in the list of strings related to the first user search query.