Entity-Based Query Filtering for Search Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional querying methods for social networks in job searches often result in irrelevant results due to keyword-based searching, leading to low precision and recall, making it difficult for recruiters to find suitable candidates.
Innovation Solution
The system uses a standardized entity taxonomy to identify and tag entities in search queries, allowing for semantic query representation, query rewriting, and result filtering to ensure that only relevant results are retrieved, based on confidence scores and entity relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If keyword-based searching is used to increase recall, then more results are retrieved, but precision deteriorates due to irrelevant results
Solution Approach 1:
The patent introduces an intermediary processing layer between the keyword search and result delivery. This layer includes entity recognition modules that identify named entities in search queries, entity relationship graphs that model semantic connections, and query rewriting mechanisms that transform simple keyword searches into structured entity-based queries. This intermediary processing enables the system to understand the intent behind keywords and filter results based on semantic relevance rather than just keyword matching.
Solution Approach 2:
The patent changes the search parameters from simple keyword matching to multi-dimensional entity-based parameters. Instead of searching solely by keywords, the system transforms queries into structured representations that include entity types, relationships, and semantic constraints. This parameter transformation allows the search to operate on multiple dimensions simultaneously, improving precision while maintaining recall through the entity relationship graph that captures semantic connections.
2Quantity of substance
If multiple keywords are used in search queries, then recall improves by capturing more potential results, but precision worsens due to inclusion of irrelevant results
Solution Approach 1:
The patent segments the search query into distinct entity components rather than treating it as a single keyword string. Each keyword is identified and classified as a specific entity type (e.g., skill, company, location, education). The system then segments the search process into multiple stages: entity recognition, entity relationship modeling, query rewriting, and filtered retrieval. This segmentation allows precise control over how each entity contributes to the search, enabling the system to maintain high recall while improving precision through structured entity-based filtering.
3Ease of operation
If traditional keyword searching is used, then ease of operation is maintained, but measurement precision of search results deteriorates
Solution Approach 1:
The patent implements self-service functionality where the search system automatically performs entity recognition, entity relationship analysis, and query rewriting without requiring user intervention. The system self-adapts to understand search intent by analyzing entity patterns and relationships in the data. This self-service approach maintains ease of operation for users while dramatically improving search precision through automated semantic processing and entity-based filtering.
Data Source
AI summary
In an example embodiment, one or more query terms are obtained. For each of the one or more query terms, a standardized entity taxonomy is searched to locate a standardized entity that most closely matches the query term. A confidence score is calculated for the query term-standardized entity pair for the standardized entity that most closely matches the query term. In response to a determination that the confidence score transgresses a threshold, the query term is associated with an entity identification corresponding to the standardized entity that most closely matches the query term. One or more query rewriting rules corresponding to an entity type of the standardized entity having the entity identification are obtained. The one or more query rewriting rules are executed to rewrite the first query such that the rewritten query, when performed on a data source, returns fewer search results than the first query would have.