Link enrichment for diffusion-based graph node kernels

Dinh Tran-Van; Alessandro Sperduti; Fabrizio Costa

doi:10.1007/978-3-319-68612-7_18

Link enrichment for diffusion-based graph node kernels

Dinh Tran-Van, Alessandro Sperduti, Fabrizio Costa^*

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

The notion of node similarity is key in many graph processing techniques and it is especially important in diffusion graph kernels. However, when the graph structure is affected by noise in the form of missing links, similarities are distorted proportionally to the sparsity of the graph and to the fraction of missing links. Here, we introduce the notion of link enrichment, that is, performing link prediction in order to improve the performance of diffusion-based kernels. We empirically show a robust and large effect for the combination of a number of link prediction and a number of diffusion kernel techniques on several gene-disease association problems.

Original language	English
Title of host publication	Artificial Neural Networks and Machine Learning – ICANN 2017 - 26th International Conference on Artificial Neural Networks, Proceedings
Editors	Alessandra Lintas, Alessandro E. Villa, Stefano Rovetta, Paul F. Verschure
Publisher	Springer Verlag
Pages	155-162
Number of pages	8
ISBN (Print)	9783319686110
DOIs	https://doi.org/10.1007/978-3-319-68612-7_18
State	Published - 2017
Externally published	Yes
Event	26th International Conference on Artificial Neural Networks, ICANN 2017 - Alghero, Italy Duration: 11 Sep 2017 → 14 Sep 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10614 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	26th International Conference on Artificial Neural Networks, ICANN 2017
Country/Territory	Italy
City	Alghero
Period	11/09/17 → 14/09/17

Bibliographical note

Publisher Copyright:
© Springer International Publishing AG 2017.

Keywords

Diffusion kernels
Graph kernels
Link prediction

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-319-68612-7_18

Cite this

Tran-Van, D., Sperduti, A., & Costa, F. (2017). Link enrichment for diffusion-based graph node kernels. In A. Lintas, A. E. Villa, S. Rovetta, & P. F. Verschure (Eds.), Artificial Neural Networks and Machine Learning – ICANN 2017 - 26th International Conference on Artificial Neural Networks, Proceedings (pp. 155-162). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10614 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-68612-7_18

Tran-Van, Dinh ; Sperduti, Alessandro ; Costa, Fabrizio. / Link enrichment for diffusion-based graph node kernels. Artificial Neural Networks and Machine Learning – ICANN 2017 - 26th International Conference on Artificial Neural Networks, Proceedings. editor / Alessandra Lintas ; Alessandro E. Villa ; Stefano Rovetta ; Paul F. Verschure. Springer Verlag, 2017. pp. 155-162 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ddb8d842f69a43a8b389674733a5a62a,

title = "Link enrichment for diffusion-based graph node kernels",

abstract = "The notion of node similarity is key in many graph processing techniques and it is especially important in diffusion graph kernels. However, when the graph structure is affected by noise in the form of missing links, similarities are distorted proportionally to the sparsity of the graph and to the fraction of missing links. Here, we introduce the notion of link enrichment, that is, performing link prediction in order to improve the performance of diffusion-based kernels. We empirically show a robust and large effect for the combination of a number of link prediction and a number of diffusion kernel techniques on several gene-disease association problems.",

keywords = "Diffusion kernels, Graph kernels, Link prediction",

author = "Dinh Tran-Van and Alessandro Sperduti and Fabrizio Costa",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 26th International Conference on Artificial Neural Networks, ICANN 2017 ; Conference date: 11-09-2017 Through 14-09-2017",

year = "2017",

doi = "10.1007/978-3-319-68612-7\_18",

language = "English",

isbn = "9783319686110",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "155--162",

editor = "Alessandra Lintas and Villa, \{Alessandro E.\} and Stefano Rovetta and Verschure, \{Paul F.\}",

booktitle = "Artificial Neural Networks and Machine Learning – ICANN 2017 - 26th International Conference on Artificial Neural Networks, Proceedings",

address = "Germany",

}

Tran-Van, D, Sperduti, A & Costa, F 2017, Link enrichment for diffusion-based graph node kernels. in A Lintas, AE Villa, S Rovetta & PF Verschure (eds), Artificial Neural Networks and Machine Learning – ICANN 2017 - 26th International Conference on Artificial Neural Networks, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10614 LNCS, Springer Verlag, pp. 155-162, 26th International Conference on Artificial Neural Networks, ICANN 2017, Alghero, Italy, 11/09/17. https://doi.org/10.1007/978-3-319-68612-7_18

Link enrichment for diffusion-based graph node kernels. / Tran-Van, Dinh; Sperduti, Alessandro; Costa, Fabrizio.
Artificial Neural Networks and Machine Learning – ICANN 2017 - 26th International Conference on Artificial Neural Networks, Proceedings. ed. / Alessandra Lintas; Alessandro E. Villa; Stefano Rovetta; Paul F. Verschure. Springer Verlag, 2017. p. 155-162 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10614 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Link enrichment for diffusion-based graph node kernels

AU - Tran-Van, Dinh

AU - Sperduti, Alessandro

AU - Costa, Fabrizio

N1 - Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - The notion of node similarity is key in many graph processing techniques and it is especially important in diffusion graph kernels. However, when the graph structure is affected by noise in the form of missing links, similarities are distorted proportionally to the sparsity of the graph and to the fraction of missing links. Here, we introduce the notion of link enrichment, that is, performing link prediction in order to improve the performance of diffusion-based kernels. We empirically show a robust and large effect for the combination of a number of link prediction and a number of diffusion kernel techniques on several gene-disease association problems.

AB - The notion of node similarity is key in many graph processing techniques and it is especially important in diffusion graph kernels. However, when the graph structure is affected by noise in the form of missing links, similarities are distorted proportionally to the sparsity of the graph and to the fraction of missing links. Here, we introduce the notion of link enrichment, that is, performing link prediction in order to improve the performance of diffusion-based kernels. We empirically show a robust and large effect for the combination of a number of link prediction and a number of diffusion kernel techniques on several gene-disease association problems.

KW - Diffusion kernels

KW - Graph kernels

KW - Link prediction

UR - https://www.scopus.com/pages/publications/85034263546

U2 - 10.1007/978-3-319-68612-7_18

DO - 10.1007/978-3-319-68612-7_18

M3 - Conference contribution

AN - SCOPUS:85034263546

SN - 9783319686110

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 155

EP - 162

BT - Artificial Neural Networks and Machine Learning – ICANN 2017 - 26th International Conference on Artificial Neural Networks, Proceedings

A2 - Lintas, Alessandra

A2 - Villa, Alessandro E.

A2 - Rovetta, Stefano

A2 - Verschure, Paul F.

PB - Springer Verlag

T2 - 26th International Conference on Artificial Neural Networks, ICANN 2017

Y2 - 11 September 2017 through 14 September 2017

ER -

Tran-Van D, Sperduti A, Costa F. Link enrichment for diffusion-based graph node kernels. In Lintas A, Villa AE, Rovetta S, Verschure PF, editors, Artificial Neural Networks and Machine Learning – ICANN 2017 - 26th International Conference on Artificial Neural Networks, Proceedings. Springer Verlag. 2017. p. 155-162. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-68612-7_18

Link enrichment for diffusion-based graph node kernels

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this