AI Semantic Extraction for Database Ontology Construction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing database technologies face challenges in extracting semantic information from runtime behaviors, which is crucial for establishing ontology and improving computational efficiency, as semantic variables are often buried and difficult to identify due to different names across data tables.

Innovation Solution

An artificially intelligent method that monitors database information sources to identify primary semantic information, reformats it, analyzes it to establish secondary semantic information, and constructs ontologies from primary and secondary information, using techniques such as comparing data formats and values to infer relationships among variables.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If semantic information is extracted from database runtime behaviors, then computational efficiency is improved and ontology is established, but the complexity of information processing increases

Engineering Contradiction:

Improvecomputational efficiencyVSAvoidinformation processing complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the complex task of semantic information extraction into multiple distinct modules: a monitoring module that captures runtime behaviors, a reformatting module that standardizes data formats, and an analysis module that derives semantic information. This segmentation reduces processing complexity by handling different aspects separately while maintaining overall efficiency.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary reformatting layer that transforms raw runtime behavior data into a standardized format before analysis. This intermediary step acts as a mediator between the complex raw data and the analysis process, simplifying the overall information processing while enabling efficient semantic extraction.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If semantic information is automatically extracted from runtime behaviors, then ontology establishment is enhanced, but the time required for information extraction increases

Engineering Contradiction:

Improveontology establishment qualityVSAvoidinformation extraction time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary reformatting of runtime behavior data into standardized formats before the actual semantic analysis occurs. This preliminary action prepares the data in advance, making the subsequent semantic extraction faster and more reliable, thereby reducing the overall time required while maintaining high ontology quality.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces manual semantic information extraction with an automated artificial intelligence system that monitors runtime behaviors and automatically derives semantic information. This substitution of automated intelligent processing for manual methods significantly reduces extraction time while improving the reliability and consistency of ontology establishment.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Loss of information

If multiple information sources are monitored to identify semantic information, then the completeness of semantic variables is improved, but the complexity of monitoring increases

Engineering Contradiction:

Improvesemantic variable completenessVSAvoidmonitoring system complexity

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The patent implements a universal monitoring approach where the monitoring module captures runtime behaviors from multiple information sources using a unified method. This multi-functional monitoring system handles diverse sources (queries, transactions, metadata) through a single integrated process, improving semantic variable completeness without proportionally increasing monitoring complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent merges multiple information sources and their respective runtime behaviors into a unified data stream that is processed through a single reformatting and analysis pipeline. By combining multiple sources into one integrated processing flow, the system achieves complete semantic variable extraction while avoiding the complexity of managing separate processing paths for each source.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS11386084B2Systems and methods for deriving database semantic information using artificial intelligence

Publication Date: 2022.07.12 UNISYS CORP

AI summary

An artificially intelligent method includes the steps of monitoring, by a processor, information sources to identify primary semantic information; capturing, by the processor, the primary semantic information; reformatting, by the processor, the primary semantic information according to a predetermined format; analyzing, by the processor, the primary semantic information to establish secondary semantic information; and establishing, by the processor, ontologies from the primary, secondary, and additional secondary semantic information.