Automated Database Generation for Fact Lookup Queries

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current search engines using natural language processing to answer fact lookup queries are suboptimal due to non-conforming user queries and inability to distinguish fact lookup queries from other types, often providing irrelevant results as most queries are not in natural language format and are not specifically fact lookup queries.

Innovation Solution

Automatically generating a relational database by analyzing source code of web pages to extract and infer attribute identities and values, and using query logs to identify entity identities, allowing for the creation of a database that can provide instant answers to fact lookup queries by indexing attribute values by entity identities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If natural language processing is used to answer fact lookup queries, then the system can handle queries in natural language format, but it fails when users submit queries that do not conform to natural language processing formats

Engineering Contradiction:

Improveability to handle natural language queriesVSAvoidaccuracy of query answering

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces an intermediary component that translates user queries into structured database queries. Instead of relying solely on natural language processing, the system uses a query translation layer that converts various query formats into standardized SQL queries that can be executed against the structured database, thereby maintaining reliability while accepting diverse input formats

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system changes the parameter of query processing from direct natural language interpretation to structured query translation. By transforming the query handling approach from NLP-based to database-query-based, the system achieves more reliable and accurate fact lookup results while still accepting natural language input

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If the system tries to answer every query as a fact lookup query, then it can provide specific factual answers, but it provides irrelevant results for navigational, transactional, or informational queries

Engineering Contradiction:

Improveprecision of fact lookup answersVSAvoidability to handle different query types

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent segments the query processing system into distinct modules: a query type classification module that identifies whether a query is fact lookup, navigational, transactional, or informational, and a specialized processing module for each query type. This segmentation allows the system to apply fact lookup processing only when appropriate, avoiding irrelevant results for other query types

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system dynamically adjusts its processing approach based on the identified query type. The query router selectively applies fact lookup processing only to queries classified as fact lookup queries, while directing other query types to appropriate handling mechanisms, thereby maintaining both precision and adaptability

Inventive Principle:

Principle #15Dynamics

3Quantity of substance

If a search engine provides multiple web pages as results, then it can provide comprehensive information, but it fails to provide a single correct answer for fact lookup queries

Engineering Contradiction:

Improveamount of information providedVSAvoidclarity of factual answer

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent extracts the specific factual answer from multiple potential sources by querying the structured database with precise entity-attribute queries. The system extracts only the relevant factual information needed to answer the specific query, filtering out unnecessary web pages while maintaining answer accuracy and clarity

Inventive Principle:

Principle #2Taking out (Extraction)

4Device complexity

If the system uses natural language processing for all queries, then it can maintain a simple processing pipeline, but it increases computational complexity and reduces efficiency for non-natural language queries

Engineering Contradiction:

Improvesimplicity of processing pipelineVSAvoidquery processing efficiency

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The patent performs preliminary action by pre-processing and structuring data from web pages into a structured database format before queries are submitted. This preliminary structuring eliminates the need for complex natural language processing during query execution, thereby simplifying the processing pipeline and improving query processing efficiency

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9460207B2Automated database generation for answering fact lookup queries

Publication Date: 2016.10.04 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Technologies pertaining to fact lookup queries are described herein. A relational database is automatically built by extracting attribute identities and attribute values from a one dimensional table, wherein the one dimensional table does not include an entity identity that corresponds to the attribute identity and the attribute value. The entity identity is inferred, and the attribute value is indexed in a relational database by the entity identity and the attribute identity. When a query is issued by a user that includes the entity identity and the attribute identity, the corresponding attribute value in the relational database is returned to the user.