Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
2 files | 514.50 KB
Add to list
  1. Search results
  2. Walk extraction strategies for node e...

Walk extraction strategies for node embeddings with RDF2Vec in knowledge graphs

Bram Steenwinckel (UGent) , Gilles Vandewiele (UGent) , Pieter Bonte (UGent) , Michael Weyns (UGent) , Heiko Paulheim, Petar Ristoski, Filip De Turck (UGent) and Femke Ongenae (UGent)
(2021) DATABASE AND EXPERT SYSTEMS APPLICATIONS - DEXA 2021 WORKSHOPS. In Communications in Computer and Information Science 1479. p.70-80
Author
Organization
Abstract
As Knowledge Graphs are symbolic constructs, specialized techniques have to be applied in order to make them compatible with data mining techniques. RDF2Vec is an unsupervised technique that can create task-agnostic numerical representations of the nodes in a KG by extending successful language modeling techniques. The original work proposed the Weisfeiler-Lehman kernel to improve the quality of the representations. However, in this work, we show that the Weisfeiler-Lehman kernel does little to improve walk embeddings in the context of a single Knowledge Graph. As an alternative, we examined five alternative strategies to extract information complementary to basic random walks and compare them on several benchmark datasets to show that research within this field is still relevant for node classification tasks.
Keywords
Knowledge graphs, Embeddings, Representation learning

Downloads

  • Download
    7993 acc.pdf
    • full text (Accepted manuscript)
    • |
    • open access
    • |
    • PDF
    • |
    • 315.71 KB
  • Download
    (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 198.79 KB

Citation

Please use this url to cite or link to this publication:

MLA
Steenwinckel, Bram, et al. “Walk Extraction Strategies for Node Embeddings with RDF2Vec in Knowledge Graphs.” DATABASE AND EXPERT SYSTEMS APPLICATIONS - DEXA 2021 WORKSHOPS, edited by G. Kotsis et al., vol. 1479, Springer, 2021, pp. 70–80, doi:10.1007/978-3-030-87101-7_8.
APA
Steenwinckel, B., Vandewiele, G., Bonte, P., Weyns, M., Paulheim, H., Ristoski, P., … Ongenae, F. (2021). Walk extraction strategies for node embeddings with RDF2Vec in knowledge graphs. In G. Kotsis, A. M. Tjoa, I. Khalil, B. Moser, A. Mashkoor, J. Sametinger, … S. Khan (Eds.), DATABASE AND EXPERT SYSTEMS APPLICATIONS - DEXA 2021 WORKSHOPS (Vol. 1479, pp. 70–80). https://doi.org/10.1007/978-3-030-87101-7_8
Chicago author-date
Steenwinckel, Bram, Gilles Vandewiele, Pieter Bonte, Michael Weyns, Heiko Paulheim, Petar Ristoski, Filip De Turck, and Femke Ongenae. 2021. “Walk Extraction Strategies for Node Embeddings with RDF2Vec in Knowledge Graphs.” In DATABASE AND EXPERT SYSTEMS APPLICATIONS - DEXA 2021 WORKSHOPS, edited by G. Kotsis, A. M. Tjoa, I. Khalil, B. Moser, A. Mashkoor, J. Sametinger, A. Fensel, et al., 1479:70–80. Springer. https://doi.org/10.1007/978-3-030-87101-7_8.
Chicago author-date (all authors)
Steenwinckel, Bram, Gilles Vandewiele, Pieter Bonte, Michael Weyns, Heiko Paulheim, Petar Ristoski, Filip De Turck, and Femke Ongenae. 2021. “Walk Extraction Strategies for Node Embeddings with RDF2Vec in Knowledge Graphs.” In DATABASE AND EXPERT SYSTEMS APPLICATIONS - DEXA 2021 WORKSHOPS, ed by. G. Kotsis, A. M. Tjoa, I. Khalil, B. Moser, A. Mashkoor, J. Sametinger, A. Fensel, J. Martinez-Gil, L. Fischer, G. Czech, F. Sobieczky, and S. Khan, 1479:70–80. Springer. doi:10.1007/978-3-030-87101-7_8.
Vancouver
1.
Steenwinckel B, Vandewiele G, Bonte P, Weyns M, Paulheim H, Ristoski P, et al. Walk extraction strategies for node embeddings with RDF2Vec in knowledge graphs. In: Kotsis G, Tjoa AM, Khalil I, Moser B, Mashkoor A, Sametinger J, et al., editors. DATABASE AND EXPERT SYSTEMS APPLICATIONS - DEXA 2021 WORKSHOPS. Springer; 2021. p. 70–80.
IEEE
[1]
B. Steenwinckel et al., “Walk extraction strategies for node embeddings with RDF2Vec in knowledge graphs,” in DATABASE AND EXPERT SYSTEMS APPLICATIONS - DEXA 2021 WORKSHOPS, online, 2021, vol. 1479, pp. 70–80.
@inproceedings{8724157,
  abstract     = {{As Knowledge Graphs are symbolic constructs, specialized techniques have to be applied in order to make them compatible with data mining techniques. RDF2Vec is an unsupervised technique that can create task-agnostic numerical representations of the nodes in a KG by extending successful language modeling techniques. The original work proposed the Weisfeiler-Lehman kernel to improve the quality of the representations. However, in this work, we show that the Weisfeiler-Lehman kernel does little to improve walk embeddings in the context of a single Knowledge Graph. As an alternative, we examined five alternative strategies to extract information complementary to basic random walks and compare them on several benchmark datasets to show that research within this field is still relevant for node classification tasks.}},
  author       = {{Steenwinckel, Bram and Vandewiele, Gilles and Bonte, Pieter and Weyns, Michael and Paulheim, Heiko and Ristoski, Petar and De Turck, Filip and Ongenae, Femke}},
  booktitle    = {{DATABASE AND EXPERT SYSTEMS APPLICATIONS - DEXA 2021 WORKSHOPS}},
  editor       = {{Kotsis, G. and Tjoa, A. M. and Khalil, I. and Moser, B. and Mashkoor, A. and Sametinger, J. and Fensel, A. and Martinez-Gil, J. and Fischer, L. and Czech, G. and Sobieczky, F. and Khan, S.}},
  isbn         = {{9783030871000}},
  issn         = {{1865-0929}},
  keywords     = {{Knowledge graphs,Embeddings,Representation learning}},
  language     = {{eng}},
  location     = {{online}},
  pages        = {{70--80}},
  publisher    = {{Springer}},
  title        = {{Walk extraction strategies for node embeddings with RDF2Vec in knowledge graphs}},
  url          = {{http://dx.doi.org/10.1007/978-3-030-87101-7_8}},
  volume       = {{1479}},
  year         = {{2021}},
}
Altmetric
View in Altmetric
Web of Science
Times cited: