Anaphora Resolution in Search Query Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Search engines face challenges in properly processing follow-up queries that contain anaphora, leading to suboptimal search results due to unclear or ambiguous queries.
Innovation Solution
A system that analyzes follow-up queries for anaphora and rewrites them using knowledge gained from the initial query by replacing anaphora with associated entities, sending modified queries to a query-ranking engine for improved search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the search engine processes follow-up queries containing anaphora directly without rewriting, then the processing speed is maintained, but the search result accuracy deteriorates due to ambiguous queries
Solution Approach 1:
The system performs preliminary action by rewriting anaphora in follow-up queries before they are processed by the search engine. The anaphora resolution module analyzes the follow-up query, identifies anaphora references, replaces them with appropriate entities from the initial query context, and generates rewritten queries that are then sent to the search engine. This preliminary rewriting step ensures accurate search results while maintaining a relatively simple overall system architecture.
2Reliability
If the system rewrites follow-up queries by replacing anaphora with entities, then the query clarity is improved, but the processing time increases due to additional analysis steps
Solution Approach 1:
The system implements feedback by using the initial query and its results to inform the processing of follow-up queries. The anaphora resolution module leverages entities and context from the initial query to accurately resolve anaphora references in subsequent queries. This feedback mechanism improves query interpretation reliability while minimizing additional processing time by reusing information from the initial query analysis.
3Loss of information
If the search engine handles ambiguous anaphoric queries, then the system simplicity is maintained, but the information retrieval quality deteriorates
Solution Approach 1:
The system introduces an intermediary component - the anaphora resolution module - that sits between the user's follow-up query and the search engine. This intermediary analyzes the query, resolves anaphora references by consulting the initial query context, and generates rewritten queries with clear entity references. This intermediary layer significantly improves information retrieval quality by eliminating ambiguity while adding only moderate system complexity through a dedicated modular component.
Data Source
AI summary
The specification relates to a method of receiving a first query and a second query. The method analyzes the second query for a presence of anaphora. If anaphora is present, the method analyzes the first query for a presence of an entity that can be associated with the anaphora. If the analysis analyzing the first query returns two or more associated entities, the method forms a third query wherein the anaphora of the second query is replaced with one of the associated entities and forms a fourth query wherein the anaphora is replaced with the other of the associated entities. The third query and the fourth query are sent to a query-ranking engine. The third query and the fourth query receive a ranking and the higher-ranked query is sent to a search engine.