Knowledge Graph Generation via Adverbial Relation Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing knowledge graph technologies fail to address knowledge conflicts and do not process deep semantic connotations, leading to incomplete and inaccurate expression of knowledge.
Innovation Solution
A system and method for generating knowledge graphs that perform word segmentation and part-of-speech tagging to identify subjects, adverbs, and verbs, forming triples with adverb types as relations, which are used to create organized knowledge graphs that can efficiently and accurately connect information and facilitate deeper semantic reasoning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing semantic analysis methods are used to construct knowledge graphs, then the construction process is simple and direct, but the semantic connotation is superficial and knowledge conflicts cannot be resolved
Solution Approach 1:
The patent segments the semantic analysis process into multiple hierarchical levels: first-level extraction of basic subject-verb-object triples, second-level analysis of adverbial modifiers and their types, and third-level resolution of knowledge conflicts. This segmentation allows deep semantic analysis to be performed in manageable stages, improving analysis depth while controlling complexity through systematic decomposition of the processing pipeline.
Solution Approach 2:
The patent adds a new dimension to traditional triple extraction by incorporating adverb type classification as an additional layer of semantic analysis. Instead of merely extracting (subject, verb, object) triples, the system enriches them with adverbial information and conflict resolution markers, transforming flat triple extraction into multi-dimensional semantic representation that captures deeper meaning while maintaining structured organization.
2Reliability
If traditional triple extraction is used, then the knowledge graph construction is straightforward, but knowledge conflicts cannot be addressed and knowledge expression accuracy is reduced
Solution Approach 1:
The patent performs preliminary classification of adverbs into specific types (time, place, manner, degree, etc.) during the extraction phase, before conflict resolution is needed. This preliminary action prepares the data structure in advance with organized semantic information, enabling more accurate conflict detection and resolution later without requiring complex reprocessing, thus improving knowledge expression accuracy while managing complexity through advance preparation.
Solution Approach 2:
The patent introduces adverb type classification as an intermediary layer between raw text extraction and final knowledge graph construction. This intermediary step processes adverbial modifiers and categorizes them, serving as a bridge that enriches basic triples with semantic nuance and provides additional context for resolving knowledge conflicts, thereby improving reliability without directly increasing the complexity of the core extraction mechanism.
3Productivity
If shallow semantic analysis is performed, then the processing speed is fast, but deep semantic connotations cannot be captured and knowledge reasoning is limited
Solution Approach 1:
The patent segments semantic analysis into hierarchical stages where basic triple extraction maintains high processing speed, while optional deeper analysis of adverbial modifiers and conflict resolution provides enhanced semantic understanding when needed. This segmentation allows the system to operate efficiently at the basic level while capturing deep semantic connotations through additional processing layers, thus maintaining productivity while reducing information loss through progressive analysis.
Solution Approach 2:
The patent implements partial deep semantic analysis by focusing computational resources on analyzing adverbial modifiers and resolving conflicts only where necessary, rather than performing exhaustive analysis on all triples. This partial action approach captures essential deep semantic connotations that improve knowledge reasoning while maintaining overall processing efficiency by avoiding unnecessary excessive analysis on already clear-cut cases.
Data Source
AI summary
A method of generating knowledge graph, performed by a processing device, includes: obtaining a knowledge document, performing word segmentation and part-of-speech tagging on the knowledge document to generate a number of tagged words, obtaining a number of sentences from the tagged words according to a default sentence pattern, wherein each of the sentences includes a subject, an adverb, a verb and an object, and the adverb corresponding to an adverb type, for each of the sentences, performing: using the subject as a first entity of a triple, using the object as a second entity of the triple, and using the adverb type and the verb as a relation in the triple, and forming a knowledge graph using the triple corresponding to each of the sentences.