AI Research Answer

https://doi.org/10.1111/exsy.70026

5 cited papers · May 1, 2026 · Powered by Researchly AI

🧠

TL;DR

The user has submitted a DOI link rather than a research question. The retrieved evidence blocks do not contain a paper matching the DOI `10.1111/exsy.70026`, n…

The user has submitted a DOI link rather than a research question. The retrieved evidence blocks do not contain a paper matching the DOI

code

10.1111/exsy.70026

, nor do they provide sufficient information to describe its contents. I cannot support a factual summary of that specific paper from the retrieved papers.

However, the evidence blocks do cover related themes — knowledge graphs, ontology alignment, and semantic knowledge representation — which I can address in a research context.¹²

1

A Survey on Knowledge Graphs: Representation, Acquisition, and ApplicationsShaoxiong Ji, Shirui Pan et al.2021IEEE Transactions on Neural Networks and Learning Systems

2

Integration of Contextual Descriptors in Ontology Alignment for Enrichment of Semantic CorrespondenceEduard Manziuk, Oleksander Barmak et al.2024arXiv

Knowledge graphs (KGs) have emerged as a major research direction for representing structured relational knowledge between entities, supporting tasks such as node classification, link prediction, and knowledge-aware applications.¹

Ji et al. (2021)

Ontology alignment — finding semantic correspondences between entities across ontologies — is a key interoperability challenge in this space.²

Efeoglu (2024)

Knowledge Graph (KG) — A graph-structured data model representing entities and their relations, enabling representation learning, knowledge acquisition, and downstream reasoning tasks.

¹²

Ji et al. (2021)

1

A Survey on Knowledge Graphs: Representation, Acquisition, and ApplicationsShaoxiong Ji, Shirui Pan et al.2021IEEE Transactions on Neural Networks and Learning Systems

2

Improving Graph Embeddings in Machine Learning Using Knowledge Completion with Validation in a Case Study on COVID-19 SpreadRosario Napoli, Gabriele Morabito et al.2025arXiv

Ontology Alignment — The process of identifying relationships or correspondences between semantically similar entities across two or more ontologies to resolve interoperability problems.

³

Efeoglu (2024)

3

GraphMatcher: A Graph Representation Learning Approach for Ontology MatchingSefika Efeoglu2024arXiv

Knowledge Completion — A phase in graph machine learning pipelines that uncovers latent or implicit knowledge hidden in sparse datasets, reshaping graph topology before embedding generation.

²Napoli et al. (2025)²²²

Contextual Descriptors — Enrichment mechanisms integrated into ontology alignment to capture the contextual dependence of concepts, improving alignment metrics especially in domains such as privacy, responsibility, and autonomy.

⁴Manziuk et al. (2024)⁴⁴⁴

4

Integration of Contextual Descriptors in Ontology Alignment for Enrichment of Semantic CorrespondenceEduard Manziuk, Oleksander Barmak et al.2024arXiv

Want to research your own topic? Try it free →

Diagram

Raw Knowledge Sources
 |
 v
+-------------------+
| Knowledge Elicitation |
| & Formalization |
+-------------------+
 |
 v
+-------------------+
| Ontology / KG |
| Construction |
+-------------------+
 |
 v
+-------------------+
| Knowledge |
| Completion (KC) | <-- Transitive/decay-based inference
+-------------------+
 |
 v
+-------------------+
| Graph Embedding |
| (GraphSAGE / |
| Node2Vec) |
+-------------------+
 |
 v
+-------------------+
| Downstream Tasks |
| (Link Prediction, |
| Node Classification,|
| Ontology Matching)|
+-------------------+

Ontology Alignment approaches have evolved from simple string-matching to graph-based and contextual methods.¹The GraphMatcher system uses a graph attention mechanism to compute higher-level representations of ontology classes together with their surrounding terms, achieving notable results on the OAEI 2022 conference track benchmark.¹

Efeoglu (2024)

1

GraphMatcher: A Graph Representation Learning Approach for Ontology MatchingSefika Efeoglu2024arXiv

The integration of contextual descriptors into ontology alignment yielded an average overall improvement of approximately 4.36% in alignment metrics, with the strongest gains in the areas of privacy, responsibility, and freedom & autonomy.²

Manziuk et al. (2024)

2

Integration of Contextual Descriptors in Ontology Alignment for Enrichment of Semantic CorrespondenceEduard Manziuk, Oleksander Barmak et al.2024arXiv

In graph machine learning, introducing a Knowledge Completion phase before embedding generation was shown to significantly alter embedding space geometry in GraphSAGE and Node2Vec models, demonstrating that KC is not merely an enrichment step but a transformative one that redefines representation quality.³

Napoli et al. (2025)

3

Improving Graph Embeddings in Machine Learning Using Knowledge Completion with Validation in a Case Study on COVID-19 SpreadRosario Napoli, Gabriele Morabito et al.2025arXiv

Want to research your own topic? Try it free →

The evidence blocks do not contain the paper at DOI

code

10.1111/exsy.70026

, so no claims about that specific work can be supported.¹Additionally, knowledge graph refinement remains an open challenge — early KG construction methods produced graphs with errors and incompleteness that required dedicated refinement pipelines.²

Paulheim (2016)

1

Integration of Contextual Descriptors in Ontology Alignment for Enrichment of Semantic CorrespondenceEduard Manziuk, Oleksander Barmak et al.2024arXiv

2

Knowledge graph refinement: A survey of approaches and evaluation methodsHeiko Paulheim2016Semantic Web

Ontology alignment systems also face scalability and domain-specificity constraints, as graph attention approaches like GraphMatcher are evaluated on specific benchmark tracks (e.g., OAEI conference track) and may not generalise across all domains. Efeoglu (2024)

Knowledge graphs are a foundational structure for representing relational knowledge, supporting a wide range of AI and reasoning applications.

¹Ji et al. (2021)

Contextual descriptors provide measurable improvements (~4.36% average) in ontology alignment quality, particularly for ethically complex domains.

²Manziuk et al. (2024)

Knowledge Completion before graph embedding is a transformative — not merely additive — step that reshapes embedding space geometry.

³Napoli et al. (2025)

Graph attention-based ontology matching (GraphMatcher) achieves strong benchmark results on OAEI 2022, advancing beyond purely lexical alignment methods.

⁴Efeoglu (2024)

KG refinement remains an active research challenge, as real-world knowledge graphs contain errors and incompleteness requiring dedicated correction approaches.

⁵

Paulheim (2016)

1

A Survey on Knowledge Graphs: Representation, Acquisition, and ApplicationsShaoxiong Ji, Shirui Pan et al.2021IEEE Transactions on Neural Networks and Learning Systems

2

Integration of Contextual Descriptors in Ontology Alignment for Enrichment of Semantic CorrespondenceEduard Manziuk, Oleksander Barmak et al.2024arXiv

3

Improving Graph Embeddings in Machine Learning Using Knowledge Completion with Validation in a Case Study on COVID-19 SpreadRosario Napoli, Gabriele Morabito et al.2025arXiv

4

GraphMatcher: A Graph Representation Learning Approach for Ontology MatchingSefika Efeoglu2024arXiv

5

Knowledge graph refinement: A survey of approaches and evaluation methodsHeiko Paulheim2016Semantic Web

Want to research your own topic? Try it free →

"Ontology alignment evaluation OAEI 2023 2024 benchmark results" — to find the latest competitive systems and metrics in ontology matching.
"Knowledge graph completion transitive inference link prediction survey" — to explore methods for uncovering implicit relational knowledge in sparse KGs.
"Expert Systems journal knowledge representation semantic web applications" — to locate papers published in the Expert Systems venue (where DOI prefix
code
```
10.1111/exsy
```
originates) on related topics.

Research smarter with AI-powered citations

Researchly finds and cites academic papers for any research topic in seconds. Used by students across India.

Try Free See Pricing