Knowledge Graph Construction for Ethylene Oxide Derivatives
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The fine chemical industry, particularly in ethylene oxide derivatives production, faces challenges in integrating and correlating diverse safety production information from various sources, leading to inadequate understanding and inefficient decision-making due to data dispersion and lack of unified semantic expression.
Innovation Solution
A knowledge graph construction method is proposed, involving ontology layer construction using OWL language, multi-source data extraction through top-down and bottom-up methods, and integration of natural language processing for unstructured data, to create a comprehensive knowledge system for ethylene oxide derivatives production processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data from multiple sources are collected and stored independently, then data completeness is improved, but data integration and decision-making efficiency deteriorate
Solution Approach 1:
The patent merges multiple independent data sources (DCS process data, SIS safety data, LIMS quality data, HAZOP knowledge data) into a unified knowledge graph structure. This combines the advantage of comprehensive data collection from various sources while enabling integrated analysis and improved decision-making efficiency through centralized access and correlation capabilities.
2Ease of manufacture
If traditional structural data processing tools are used, then data storage is simplified, but feature extraction and content retrieval capabilities deteriorate
Solution Approach 1:
The patent transitions from traditional flat tabular data structures to a multi-dimensional knowledge graph structure with entities, attributes, and relationships. This dimensional transformation enables rich feature extraction and semantic retrieval capabilities while maintaining data storage through standardized schemas and ontologies.
3Loss of information
If knowledge graph construction is implemented, then data integration and semantic understanding are improved, but system complexity increases
Solution Approach 1:
The patent segments the knowledge graph construction into distinct modules: data collection from multiple sources, ontology layer construction, knowledge extraction through NLP, and application layer implementation. This segmentation manages system complexity by breaking down the complex knowledge graph construction process into manageable, independently developable components with clear interfaces.
4Ease of manufacture
If unstructured data is processed using traditional methods, then processing simplicity is maintained, but knowledge extraction accuracy deteriorates
Solution Approach 1:
The patent introduces natural language processing technologies (named entity recognition, relation extraction, event extraction) as intermediary tools between unstructured data and the knowledge graph. These intermediary processing mechanisms enhance knowledge extraction accuracy from unstructured texts while maintaining relatively simple implementation through established NLP pipelines and tools.
Data Source
AI summary
The present invention belongs to the technical field of knowledge graph, and provides a knowledge graph construction method for an ethylene oxide derivatives production process. According to data types and characteristics, data sources of the ethylene oxide derivatives production process are sorted and divided into three types: structural data, unstructured data and other types of data. An ontology layer and a data layer of a knowledge graph are constructed by combining top-down and bottom-up methods. A data-driven incremental ontology modeling method is proposed to ensure the expandability of the knowledge graph. For structural knowledge extraction, the safety of original data storage is ensured by means of virtual knowledge graph, and a new mapping mechanism is proposed to realize data materialization. For unstructured knowledge extraction, an entity extraction task is realized on the basis of a BERT-BiLSTM-CRF named entity recognition model by integrating a pre-training language model BERT.