Fish Knowledge Graph Construction Using Ontology and Multi-Modal Extraction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Traditional fish information statistics and analysis rely heavily on manual operations, leading to errors and inefficiencies in processing unstructured big data, resulting in low accuracy and inability to handle large volumes of fish-related information.

Innovation Solution

An ontology-driven method and system for constructing a fish knowledge graph, utilizing pre-trained semantic analysis models to extract knowledge from unstructured text and image data, measuring information content, and integrating the results into a multi-modal knowledge graph to enhance accuracy and efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual operations are used for fish information statistics and analysis, then operation simplicity is maintained, but accuracy and productivity deteriorate due to information omission and data statistical errors

Engineering Contradiction:

Improveaccuracy of fish information collectionVSAvoidcomplexity of information processing system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces manual mechanical operations with an automated knowledge graph construction system that uses pre-trained semantic analysis models to extract knowledge from unstructured text and image data, eliminating human errors in information collection and statistical analysis

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces a knowledge graph as an intermediary structure that mediates between raw unstructured data and analytical outputs, organizing information through entities, relationships, and attributes to enable accurate processing without direct manual intervention

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If manual operations are used for fish information processing, then system complexity is kept low, but productivity and data processing capacity deteriorate due to inability to handle unstructured big data

Engineering Contradiction:

Improveefficiency of fish information processingVSAvoidcomplexity of knowledge graph system

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent employs pre-trained semantic analysis models that perform preliminary knowledge extraction and organization before final analysis, enabling the system to efficiently process large volumes of unstructured fish information data without requiring complex real-time processing

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent segments the fish information processing task into distinct modules: data collection, knowledge extraction, knowledge graph construction, and analysis, allowing each component to handle specific data types and functions independently, thereby improving overall productivity

Inventive Principle:

Principle #1Segmentation

3Reliability

If manual operations are used for fish data collection, then operational simplicity is maintained, but reliability deteriorates due to information omission and statistical errors

Engineering Contradiction:

Improveaccuracy of fish information statisticsVSAvoidautomation level of information processing

Core Design Contradiction:

ReliabilityVSExtent of automation

Solution Approach 1:

The patent substitutes manual data collection and statistical operations with automated semantic analysis models that systematically extract knowledge from unstructured text and image data, ensuring complete and accurate information capture without human error

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent implements feedback mechanisms where the knowledge graph structure guides the extraction process and validation steps ensure data quality, allowing the system to self-correct and maintain high reliability in fish information statistics

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20250356219A1Ontology-driven method and system for constructing fish knowledge graph in target region

Publication Date: 2025.11.20 GUANGDONG OCEAN UNIVERSITY

AI summary

An ontologically driven method and a system for constructing a fish knowledge graph in a target region, including: combing a fish data source in the target region, constructing a fish knowledge ontology in the target region, and collecting text data and image data; according to the fish knowledge ontology of the target region, performing knowledge extraction on the text data, and measuring information content of knowledge extraction results to obtain text information weights; according to the fish knowledge ontology in the target region, performing the knowledge extraction on the image data, and measuring information content of the knowledge extraction results to obtain image information weights; according to the text information weights and the image information weights, matching the first knowledge extraction results with the fish knowledge ontology in the target region, and constructing a multi-modal knowledge graph. The processing accuracy and efficiency of the unstructured fish information can be improved.