Advanced search
1 file | 700.41 KB Add to list

ExplaiNE: An Approach for Explaining Network Embedding-based Link Predictions

Bo Kang (UGent) , Jefrey Lijffijt (UGent) and Tijl De Bie (UGent)
(2019) arXiv.
Author
Organization
Abstract
Networks are powerful data structures, but are challenging to work with for conventional machine learning methods. Network Embedding (NE) methods attempt to resolve this by learning vector representations for the nodes, for subsequent use in downstream machine learning tasks. Link Prediction (LP) is one such downstream machine learning task that is an important use case and popular benchmark for NE methods. Unfortunately, while NE methods perform exceedingly well at this task, they are lacking in transparency as compared to simpler LP approaches. We introduce ExplaiNE, an approach to offer counterfactual explanations for NE-based LP methods, by identifying existing links in the network that explain the predicted links. ExplaiNE is applicable to a broad class of NE algorithms. An extensive empirical evaluation for the NE method `Conditional Network Embedding' in particular demonstrates its accuracy and scalability.

Downloads

  • 1904.12694.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 700.41 KB

Citation

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

MLA
Kang, Bo, Jefrey Lijffijt, and Tijl De Bie. “ExplaiNE: An Approach for Explaining Network Embedding-based Link Predictions.” arXiv 2019 : n. pag. Print.
APA
Kang, B., Lijffijt, J., & De Bie, T. (2019). ExplaiNE: An Approach for Explaining Network Embedding-based Link Predictions. arXiv.
Chicago author-date
Kang, Bo, Jefrey Lijffijt, and Tijl De Bie. 2019. “ExplaiNE: An Approach for Explaining Network Embedding-based Link Predictions.” arXiv.
Chicago author-date (all authors)
Kang, Bo, Jefrey Lijffijt, and Tijl De Bie. 2019. “ExplaiNE: An Approach for Explaining Network Embedding-based Link Predictions.” arXiv.
Vancouver
1.
Kang B, Lijffijt J, De Bie T. ExplaiNE: An Approach for Explaining Network Embedding-based Link Predictions. arXiv. 2019.
IEEE
[1]
B. Kang, J. Lijffijt, and T. De Bie, “ExplaiNE: An Approach for Explaining Network Embedding-based Link Predictions,” arXiv. 2019.
@misc{8618141,
  abstract     = {Networks are powerful data structures, but are challenging to work with for conventional machine learning methods. Network Embedding (NE) methods attempt to resolve this by learning vector representations for the nodes, for subsequent use in downstream machine learning tasks. 
Link Prediction (LP) is one such downstream machine learning task that is an important use case and popular benchmark for NE methods. Unfortunately, while NE methods perform exceedingly well at this task, they are lacking in transparency as compared to simpler LP approaches. 
We introduce ExplaiNE, an approach to offer counterfactual explanations for NE-based LP methods, by identifying existing links in the network that explain the predicted links. ExplaiNE is applicable to a broad class of NE algorithms. An extensive empirical evaluation for the NE method `Conditional Network Embedding' in particular demonstrates its accuracy and scalability.},
  author       = {Kang, Bo and Lijffijt, Jefrey and De Bie, Tijl},
  series       = {arXiv},
  title        = {ExplaiNE: An Approach for Explaining Network Embedding-based Link Predictions},
  url          = {https://arxiv.org/pdf/1904.12694.pdf},
  year         = {2019},
}