Automated Search Task Control via Parameterized Query Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search engines face inefficiencies in handling repetitive queries and factoid questions, particularly in extracting and ranking multiple answer candidates with context, due to limitations in current micro-segment search capabilities and reliance on structured databases.
Innovation Solution
The development of a system that forms a seed query based on user input, executes a parameterized query over search results to extract answer strings matching linguistic, structural, and visual patterns, and provides output to the user, reducing processor load and network bandwidth usage through domain-specific language programming and task execution systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If search engines rely on structured databases to answer factoid questions, then answer extraction is efficient, but the information is limited to database content and cannot handle time-sensitive or web-only information
Solution Approach 1:
The patent introduces an answer extraction system that acts as an intermediary between search engines and structured databases. This system executes parameterized queries over search result documents to extract answers, combining the breadth of web search with the precision of structured query processing. The intermediary layer enables flexible information sourcing while maintaining efficient extraction through predefined query templates.
Solution Approach 2:
The patent segments the answer extraction process into distinct components: seed query generation, parameterized query execution, pattern matching, and answer ranking. This segmentation allows each component to be optimized independently, with parameterized queries handling structured extraction and pattern matching handling unstructured web content, thus resolving the contradiction between efficiency and versatility.
2Measurement precision
If answer extraction code is hard-coded for each micro-segment, then extraction precision is high, but the system complexity increases and maintenance becomes difficult
Solution Approach 1:
The patent implements universal parameterized query templates that can handle multiple micro-segments through parameter substitution rather than separate hard-coded extractors for each. A single parameterized query framework serves multiple purposes by accepting different parameters (entities, attributes, relationships) to extract various types of information, reducing system complexity while maintaining precision through targeted parameterization.
Solution Approach 2:
The system uses parameter changes to adapt a single extraction template to different micro-segments. By varying parameters such as the target entity, attribute type, and relationship pattern, the same query template can precisely extract different information types without requiring separate hard-coded logic for each case, thus maintaining precision while reducing complexity.
3Reliability
If the system extracts and ranks multiple answer candidates with context, then answer quality improves, but processor load and network bandwidth usage increase
Solution Approach 1:
The patent applies partial action by extracting and ranking only a limited number of top answer candidates rather than processing all possible answers. The system extracts answers matching linguistic, structural, and visual patterns, ranks them by relevance, and returns the top N candidates with context. This partial processing maintains answer quality while significantly reducing processor load and network bandwidth compared to exhaustive extraction.
4Ease of operation
If search engines handle repetitive batch queries manually, then query flexibility is maintained, but time consumption and user effort increase
Solution Approach 1:
The patent implements preliminary action through parameterized query templates that are pre-configured with extraction logic, patterns, and ranking criteria. When handling repetitive batch queries, the system instantiates these pre-prepared templates with specific parameters rather than creating extraction logic from scratch for each query. This preliminary preparation maintains flexibility through parameter variation while dramatically reducing time consumption through template reuse.
Data Source
AI summary
Various technologies described herein pertain to controlling execution of an automated search task on search results returned by a search engine. The search results are received, where the search results are returned by the search engine responsive to the search engine receiving a seed query. An instantiation of a parameterized query for the automated search task is executed over documents specified by the search results, where the instantiation of the parameterized query describes a linguistic pattern, a structural pattern, and a visual pattern. Further, a set of answer strings is extracted from the documents. The answer strings in the set match the linguistic pattern, the structural pattern, and the visual pattern. The search engine is controlled to provide an output in response to the user search query, the output being based on the set of answer strings extracted from the documents.