Natural Language to Formal Query Conversion in Semi-Structured Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing techniques struggle to effectively convert natural language into formal language for question answering systems using semi-structured data, particularly when the data includes text with semantic structure, as they fail to utilize the semantic structure of the text correctly.

Innovation Solution

A machine learning-based information processing apparatus and method that trains a converter to accept natural language text and output formal language expressions by extracting node text from semi-structured data, generating node text expressions, calculating answers, and updating parameters to improve the likelihood of correct output.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If a semantic parser is used to convert natural language into formal language based on parameters set in advance, then the conversion can be automated, but the system cannot suitably handle diverse natural language expressions and semi-structured data with semantic structure

Engineering Contradiction:

Improveautomatic conversion of natural language to formal languageVSAvoidability to handle diverse natural language expressions and semi-structured data

Core Design Contradiction:

Extent of automationVSAdaptability or versatility

Solution Approach 1:

The patent applies preliminary action by extracting semantic structure information from semi-structured data in advance and incorporating it into the training dataset before model training. This allows the semantic parser to learn from pre-processed semantic patterns, improving its ability to handle diverse natural language expressions while maintaining automation.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes parameters by transitioning from fixed parameters set in advance to dynamic parameters learned through machine training. The semantic parser is trained using machine learning algorithms that automatically adjust parameters based on training data, enabling the system to adapt to diverse natural language expressions and semi-structured data formats.

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If machine learning is used to automatically construct a semantic parser, then the system can adapt to diverse natural language expressions, but the system cannot effectively utilize semi-structured data with semantic structure

Engineering Contradiction:

Improveability to handle diverse natural language expressionsVSAvoideffectiveness in utilizing semi-structured data with semantic structure

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent applies segmentation by dividing the processing into distinct components: extracting semantic structure from semi-structured data, converting natural language to formal language, and integrating both through machine learning. This segmented approach allows the system to reliably process semi-structured data while maintaining adaptability to diverse natural language expressions.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary mechanism that bridges semi-structured data and natural language processing. The semantic structure extracted from semi-structured data serves as an intermediary representation that guides the semantic parser in converting natural language to formal language, improving reliability in utilizing semi-structured data while maintaining adaptability.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Manufacturing precision

If manual rules are constructed for converting natural language to formal language, then the conversion can be precise, but it becomes difficult to handle diverse natural language expressions

Engineering Contradiction:

Improveprecision of natural language to formal language conversionVSAvoidcomplexity of manual rule construction

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies self-service by enabling the semantic parser to automatically learn and improve its conversion rules through machine training. Instead of requiring manual rule construction, the system trains itself using training data, automatically adapting to diverse natural language expressions while maintaining conversion precision through learned patterns.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS11669691B2Information processing apparatus, information processing method, and computer readable recording medium

Publication Date: 2023.06.06 NEC CORP

AI summary

An information processing apparatus (10) includes: a formal language query accepting unit (12) that accepts a query expression and correct answer data; a semi-structured data accepting unit (14) that accepts semi-structured data that includes text nodes; a node text extraction unit (16) that extracts natural language text from the text node, as node text; a node text expression generation unit (18) that receives the node text from the a converter (100) and obtains node text expressions; an answer calculation unit (20) that calculates an answer to the query expression with use of the node text expressions; and an update unit (22) that, if the answer calculated by the answer calculation unit (20) matches the correct answer data, updates parameters in the converter (100) such that the corresponding node text expression is more likely to be output in the converter (100).