Document Ranking via Ordered Sub-Query Series
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional document retrieval systems sacrifice response time to improve recall and precision, and further compromise response time through search term relaxation methods, leading to inefficiencies in processing cycles and bandwidth usage.
Innovation Solution
The system employs an ordered series of sub-queries that prioritize metadata string matching and content analysis to pre-select relevant documents, eliminating the need for post-retrieval relevance ranking, thereby reducing the number of documents retrieved and processing power required.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional document retrieval ranks the entire set of documents, then recall and precision are improved, but response time is sacrificed
Solution Approach 1:
The patent segments the document retrieval process into multiple stages: an initial retrieval stage that uses relaxed search terms to quickly identify candidate documents, and a subsequent ranking stage that applies precise relevance criteria only to this reduced candidate set. This segmentation allows the system to maintain both speed and accuracy by avoiding the need to rank the entire document corpus.
Solution Approach 2:
The patent applies preliminary relaxation of search terms before the actual retrieval to cast a wider net and identify potential relevant documents quickly. This preliminary action filters out obviously irrelevant documents before they enter the ranking process, reducing the workload for subsequent precise ranking operations.
2Quantity of substance
If search term relaxation is performed to improve recall, then recall is improved, but response time is further sacrificed
Solution Approach 1:
The patent separates relaxation operations from the main retrieval process, performing relaxation only on search terms before initial retrieval. This segmentation ensures that relaxation expands the search scope without requiring the system to process every possible document variation through the full ranking pipeline, thus maintaining response time.
Solution Approach 2:
The patent applies relaxation partially - only to the search query formulation stage rather than to every document processing step. This partial application of relaxation provides sufficient recall improvement without the excessive computational cost of relaxing every possible query aspect throughout the entire retrieval process.
Data Source
AI summary
Systems, methods, media, and other embodiments associated with ranking documents by providing a search engine with a series of sub-queries generated from an original query are described. One example system includes input logic for receiving a query. The example system may include a relaxation logic configured to produce sub-queries from the query. The sub-queries may describe metadata string matching, content string matching, and/or metadata numerical attribute analysis. The sub-queries may be provided by an output logic to a search engine in an order that facilitates defining document relevance without requiring post-retrieval relevance ranking.