Knowledge Graph Noise Detection via Ontology Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional noise detection techniques in knowledge graphs primarily focus on factually false noise, neglecting the classification and removal of inconsistent and generic noise, which hampers the accuracy and efficiency of knowledge graphs constructed using autonomous information extraction techniques.
Innovation Solution
A system and method for autonomously classifying and identifying various types of noise in knowledge graphs, including inconsistent, generic, and factually false noise, using a processor-based system with a knowledge extraction component that analyzes knowledge base triples through ontology mapping, natural language processing, and evidence search in external knowledge graphs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional noise detection techniques focus only on factually false noise, then the detection process is simple, but the accuracy and completeness of noise removal is insufficient
Solution Approach 1:
The patent segments noise detection into three distinct classification categories: factually false noise, inconsistent noise, and generic noise. Each category is detected using specialized techniques tailored to its characteristics, allowing comprehensive noise removal while maintaining systematic complexity management through modular detection components
2Productivity
If autonomous information extraction techniques are used to construct knowledge graphs, then the construction efficiency is improved, but noise generation increases
Solution Approach 1:
The patent converts the harmful effect of noise generation during autonomous information extraction into a benefit by developing specialized detection mechanisms that identify and remove three types of noise (factually false, inconsistent, and generic). The noise classification system transforms the problem of high noise generation into an opportunity to implement comprehensive purification, ultimately improving knowledge graph quality while preserving extraction efficiency
3Reliability
If comprehensive noise classification is implemented, then the purity of knowledge graph is improved, but the processing time increases
Solution Approach 1:
The patent segments the noise processing task into three parallel classification streams for factually false noise, inconsistent noise, and generic noise. This segmentation allows simultaneous processing of different noise types through specialized detectors, improving knowledge graph purity while minimizing time loss through efficient parallel execution of classification algorithms
Data Source
AI summary
Techniques regarding autonomous classification and/or identification of various types of noise comprised within a knowledge graph are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise a knowledge extraction component, operatively coupled to the processor, that can classify a type of noise comprised within a knowledge graph. The type of noise can be generated by an information extraction process.