Query Routing via Structured Data Annotation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing web search technologies fail to effectively handle queries with semantic meaning, often returning irrelevant results due to simple keyword matching, and often search large datasets instead of focused data sources, which may contain more relevant information.
Innovation Solution
Modifying online web search queries into expressions to access structured data stores by processing tokens against dictionaries and patterns, using an annotation mechanism to find matching patterns and generate expressions for querying structured data sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If simple keyword matching is used to search large document sets, then search coverage is comprehensive, but search relevance deteriorates due to inability to handle semantic meaning
Solution Approach 1:
The patent segments the query processing into distinct components: tokenization of the query into individual terms, classification of each token into semantic categories (e.g., entity types, attributes), and structured representation of query intent. This segmentation allows the system to handle semantic meaning systematically while maintaining comprehensive search coverage across different data sources.
Solution Approach 2:
The patent introduces an intermediary layer between the user query and the data sources. This intermediary includes a query parser that converts natural language queries into structured representations, and a result merger that combines results from multiple sources. This intermediary layer enables semantic understanding without requiring changes to the underlying data sources or search infrastructure.
2Adaptability or versatility
If queries are searched against large document sets, then comprehensive results are obtained, but performance deteriorates due to processing overhead
Solution Approach 1:
The patent implements dynamic query routing that adapts to the specific query being processed. Based on the parsed query structure and identified intent, the system dynamically determines which data sources to query and how to process the results. This dynamic approach allows the system to maintain comprehensive search coverage while optimizing performance by avoiding unnecessary processing of irrelevant data sources.
Solution Approach 2:
The patent applies partial action by selectively querying only the necessary data sources based on query analysis, rather than searching all available documents. The system performs excessive action by maintaining the capability to access comprehensive data sources when needed, while typically using a subset of sources for common query types to improve performance.
3Measurement precision
If structured data processing is implemented to improve search relevance, then query accuracy improves, but system complexity increases due to annotation and translation mechanisms
Solution Approach 1:
The patent implements a universal query processing framework that handles multiple query types and data sources through a single set of mechanisms. The annotation system uses general-purpose token classification that can be applied to different query structures, and the translation mechanisms work across various data source formats. This universality reduces the need for query-specific processing logic, thereby managing system complexity while maintaining high query accuracy.
Solution Approach 2:
The patent enables the query processing system to automatically analyze and structure queries without requiring manual intervention or complex configuration. The annotation mechanisms self-adapt to different query patterns through automated token classification, and the system automatically determines appropriate data sources and processing strategies based on the parsed query structure, reducing the operational complexity of the system.
Data Source
AI summary
Described is answering online web queries from a structured data store, such as a products database. An online web search query is modified into an evaluation expression for accessing a structured data store (e.g., a database, XML file, a flat table) to find search results. A query is matched to a pattern, which is used to generate the expression, as well as determine to which structured data store the query is routed. Tokens (e.g., words) in the query are processed against a dictionary of token classes (sets of tokens) and patterns (sets of token classes) to annotate the query to map the query to the matching pattern. A translation process generates the expression based on translation hints that correspond to the matching pattern. Also described is offline mining of query logs to generate the dictionaries.