Details
| Originalsprache | Englisch |
|---|---|
| Titel des Sammelwerks | K-CAP '23 |
| Untertitel | Proceedings of the 12th Knowledge Capture Conference 2023 |
| Seiten | 44-52 |
| Seitenumfang | 9 |
| Publikationsstatus | Veröffentlicht - 5 Dez. 2023 |
| Veranstaltung | 12th ACM International Conference on Knowledge Capture, K-CAP 2023 - Pensacola, USA / Vereinigte Staaten Dauer: 5 Dez. 2023 → 7 Dez. 2023 |
Abstract
Knowledge graphs (KGs) naturally capture the convergence of data and knowledge, making them expressive frameworks for describing and integrating heterogeneous data in a coherent and interconnected manner. However, based on the Open World Assumption (OWA), the absence of information within KGs does not indicate falsity or non-existence; it merely reflects incompleteness. Inductive learning over KGs involves predicting new relationships based on existing statements in the KG, using either numerical or symbolic learning models. The Partial Completeness Assumption (PCA) heuristic efficiently guides inductive learning methods for Link Prediction (LP) by refining predictions about absent KG relationships. Nevertheless, numeric techniques- like KG embedding models- alone may fall short in accurately predicting missing information, particularly when it comes to capturing implicit knowledge and complex relationships. We propose a hybrid method named SPaRKLE that seamlessly integrates symbolic and numerical techniques, leveraging the PCA heuristic to capture implicit knowledge and enrich KGs. We empirically compare SPaRKLE with state-of-the-art KG embedding and symbolic models, using established benchmarks. Our experimental outcomes underscore the efficacy of this hybrid approach, as it harnesses the strengths of both paradigms. SPaRKLE is publicly available on GitHub1.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Theoretische Informatik und Mathematik
- Informatik (insg.)
- Angewandte Informatik
- Informatik (insg.)
- Information systems
- Informatik (insg.)
- Software
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
K-CAP '23: Proceedings of the 12th Knowledge Capture Conference 2023. 2023. S. 44-52.
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - SPaRKLE
T2 - 12th ACM International Conference on Knowledge Capture, K-CAP 2023
AU - Purohit, Disha
AU - Chudasama, Yashrajsinh
AU - Rivas, Ariam
AU - Vidal, Maria Esther
N1 - Funding Information: This work has been supported by TrustKG - Transforming Data in Trustable Insights (GA No. P99/2020) funded by the Leibniz Association and the EraMed project P4-LUCAT (GA No. 53000015).
PY - 2023/12/5
Y1 - 2023/12/5
N2 - Knowledge graphs (KGs) naturally capture the convergence of data and knowledge, making them expressive frameworks for describing and integrating heterogeneous data in a coherent and interconnected manner. However, based on the Open World Assumption (OWA), the absence of information within KGs does not indicate falsity or non-existence; it merely reflects incompleteness. Inductive learning over KGs involves predicting new relationships based on existing statements in the KG, using either numerical or symbolic learning models. The Partial Completeness Assumption (PCA) heuristic efficiently guides inductive learning methods for Link Prediction (LP) by refining predictions about absent KG relationships. Nevertheless, numeric techniques- like KG embedding models- alone may fall short in accurately predicting missing information, particularly when it comes to capturing implicit knowledge and complex relationships. We propose a hybrid method named SPaRKLE that seamlessly integrates symbolic and numerical techniques, leveraging the PCA heuristic to capture implicit knowledge and enrich KGs. We empirically compare SPaRKLE with state-of-the-art KG embedding and symbolic models, using established benchmarks. Our experimental outcomes underscore the efficacy of this hybrid approach, as it harnesses the strengths of both paradigms. SPaRKLE is publicly available on GitHub1.
AB - Knowledge graphs (KGs) naturally capture the convergence of data and knowledge, making them expressive frameworks for describing and integrating heterogeneous data in a coherent and interconnected manner. However, based on the Open World Assumption (OWA), the absence of information within KGs does not indicate falsity or non-existence; it merely reflects incompleteness. Inductive learning over KGs involves predicting new relationships based on existing statements in the KG, using either numerical or symbolic learning models. The Partial Completeness Assumption (PCA) heuristic efficiently guides inductive learning methods for Link Prediction (LP) by refining predictions about absent KG relationships. Nevertheless, numeric techniques- like KG embedding models- alone may fall short in accurately predicting missing information, particularly when it comes to capturing implicit knowledge and complex relationships. We propose a hybrid method named SPaRKLE that seamlessly integrates symbolic and numerical techniques, leveraging the PCA heuristic to capture implicit knowledge and enrich KGs. We empirically compare SPaRKLE with state-of-the-art KG embedding and symbolic models, using established benchmarks. Our experimental outcomes underscore the efficacy of this hybrid approach, as it harnesses the strengths of both paradigms. SPaRKLE is publicly available on GitHub1.
KW - Inductive Learning
KW - Knowledge Graphs
KW - Symbolic Learning
UR - http://www.scopus.com/inward/record.url?scp=85180366887&partnerID=8YFLogxK
U2 - 10.1145/3587259.3627547
DO - 10.1145/3587259.3627547
M3 - Conference contribution
AN - SCOPUS:85180366887
SN - 979-8-4007-0141-2
SP - 44
EP - 52
BT - K-CAP '23
Y2 - 5 December 2023 through 7 December 2023
ER -