Advanced search
1 file | 661.25 KB Add to list

Network representation learning for link prediction : are we improving upon simple heuristics?

Alexandru Cristian Mara (UGent) , Jefrey Lijffijt (UGent) and Tijl De Bie (UGent)
(2020)
Author
Organization
Abstract
Network representation learning has become an active research area in recent years with many new methods showcasing their performance on downstream prediction tasks such as Link Prediction. Despite the efforts of the community to ensure reproducibility of research by providing method implementations, important issues remain. The complexity of the evaluation pipelines and abundance of design choices have led to difficulties in quantifying the progress in the field and identifying the state-of-the-art. In this work, we analyse 17 network embedding methods on 7 real-world datasets and find, using a consistent evaluation pipeline, only thin progress over the recent years. Also, many embedding methods are outperformed by simple heuristics. Finally, we discuss how standardized evaluation tools can repair this situation and boost progress in this field.
Keywords
Representation Learning, Graph Embedding, Network Embedding, Link Prediction, Evaluation

Downloads

  • preprint.pdf
    • full text (Author's original)
    • |
    • open access
    • |
    • PDF
    • |
    • 661.25 KB

Citation

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

MLA
Mara, Alexandru Cristian, et al. Network Representation Learning for Link Prediction : Are We Improving upon Simple Heuristics? 2020.
APA
Mara, A. C., Lijffijt, J., & De Bie, T. (2020). Network representation learning for link prediction : are we improving upon simple heuristics?
Chicago author-date
Mara, Alexandru Cristian, Jefrey Lijffijt, and Tijl De Bie. 2020. “Network Representation Learning for Link Prediction : Are We Improving upon Simple Heuristics?”
Chicago author-date (all authors)
Mara, Alexandru Cristian, Jefrey Lijffijt, and Tijl De Bie. 2020. “Network Representation Learning for Link Prediction : Are We Improving upon Simple Heuristics?”
Vancouver
1.
Mara AC, Lijffijt J, De Bie T. Network representation learning for link prediction : are we improving upon simple heuristics? 2020.
IEEE
[1]
A. C. Mara, J. Lijffijt, and T. De Bie, “Network representation learning for link prediction : are we improving upon simple heuristics?” 2020.
@misc{8650109,
  abstract     = {Network representation learning has become an active research area in recent years with many new methods showcasing their performance on downstream prediction tasks such as Link Prediction. Despite the efforts of the community to ensure reproducibility of research by providing method implementations, important issues remain. The complexity of the evaluation pipelines and abundance of design choices have led to difficulties in quantifying the progress in the field and identifying the state-of-the-art. In this work, we analyse 17 network embedding methods on 7 real-world datasets and find, using a consistent evaluation pipeline, only thin progress over the recent years. Also, many embedding methods are outperformed by simple heuristics. Finally, we discuss how standardized evaluation tools can repair this situation and boost progress in this field.},
  author       = {Mara, Alexandru Cristian and Lijffijt, Jefrey and De Bie, Tijl},
  keywords     = {Representation Learning,Graph Embedding,Network Embedding,Link Prediction,Evaluation},
  language     = {eng},
  title        = {Network representation learning for link prediction : are we improving upon simple heuristics?},
  url          = {https://arxiv.org/abs/2002.11522},
  year         = {2020},
}