Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

KAE: A Property-based Method for Knowledge Graph Alignment and Extension

Daqian Shi, Xiaoyue Li, Fausto Giunchiglia

TL;DR

This work tackles semantic heterogeneity in knowledge graphs by introducing a property-based, FCA-formalized framework for KG alignment and extension. It defines novel specificity-based metrics—Horizontal, Vertical, and Informational—to measure property overlap and uses them to train ML-based matchers for etype alignment and entity recognition. The approach includes a dedicated Property Matcher and two KG-extension steps that merge aligned etypes and then attach remaining entities, validated through extensive experiments and ablation studies. The results demonstrate competitive or superior performance against state-of-the-art methods and show promise for scalable, domain-agnostic KG enrichment with reduced label- and taxonomy-related errors.

Abstract

A common solution to the semantic heterogeneity problem is to perform knowledge graph (KG) extension exploiting the information encoded in one or more candidate KGs, where the alignment between the reference KG and candidate KGs is considered the critical procedure. However, existing KG alignment methods mainly rely on entity type (etype) label matching as a prerequisite, which is poorly performing in practice or not applicable in some cases. In this paper, we design a machine learning-based framework for KG extension, including an alternative novel property-based alignment approach that allows aligning etypes on the basis of the properties used to define them. The main intuition is that it is properties that intentionally define the etype, and this definition is independent of the specific label used to name an etype, and of the specific hierarchical schema of KGs. Compared with the state-of-the-art, the experimental results show the validity of the KG alignment approach and the superiority of the proposed KG extension framework, both quantitatively and qualitatively.

KAE: A Property-based Method for Knowledge Graph Alignment and Extension

TL;DR

This work tackles semantic heterogeneity in knowledge graphs by introducing a property-based, FCA-formalized framework for KG alignment and extension. It defines novel specificity-based metrics—Horizontal, Vertical, and Informational—to measure property overlap and uses them to train ML-based matchers for etype alignment and entity recognition. The approach includes a dedicated Property Matcher and two KG-extension steps that merge aligned etypes and then attach remaining entities, validated through extensive experiments and ablation studies. The results demonstrate competitive or superior performance against state-of-the-art methods and show promise for scalable, domain-agnostic KG enrichment with reduced label- and taxonomy-related errors.

Abstract

A common solution to the semantic heterogeneity problem is to perform knowledge graph (KG) extension exploiting the information encoded in one or more candidate KGs, where the alignment between the reference KG and candidate KGs is considered the critical procedure. However, existing KG alignment methods mainly rely on entity type (etype) label matching as a prerequisite, which is poorly performing in practice or not applicable in some cases. In this paper, we design a machine learning-based framework for KG extension, including an alternative novel property-based alignment approach that allows aligning etypes on the basis of the properties used to define them. The main intuition is that it is properties that intentionally define the etype, and this definition is independent of the specific label used to name an etype, and of the specific hierarchical schema of KGs. Compared with the state-of-the-art, the experimental results show the validity of the KG alignment approach and the superiority of the proposed KG extension framework, both quantitatively and qualitatively.
Paper Structure (39 sections, 7 equations, 4 figures, 8 tables, 2 algorithms)

This paper contains 39 sections, 7 equations, 4 figures, 8 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Intuitive Discussion
  3. Task Description
  4. Observations and Motivations
  5. Knowledge Graph Formalization
  6. Property-based Similarity Metrics
  7. Horizontal Specificity
  8. Vertical Specificity
  9. Informational Specificity
  10. Similarity Metrics
  11. The Proposed Method
  12. Similarity Calculation Algorithm
  13. Knowledge Graph Extension Algorithm
  14. Machine Learning-based Matchers
  15. Etype Matcher
  16. ...and 24 more sections

Figures (4)

  • Figure 1: An example of the hierarchical schema in KG.
  • Figure 2: An example of formalizing KG into FCA contexts
  • Figure 3: The framework of KG alignment and extension.
  • Figure 4: Experimental results of KG extension function.