A tree-based kernel for graphs

Giovanni Martino, Nicolò Navarin, Alessandro Sperduti

Research output: Chapter in Book/Report/Conference proceedingConference contribution

25 Citations (Scopus)

Abstract

This paper proposes a new tree-based kernel for graphs. Graphs are decomposed into multisets of ordered Directed Acyclic Graphs (DAGs) and a family of kernels computed by application of tree kernels extended to the DAG domain. We focus our attention on the efficient development of one member of this family. A technique for speeding up the computation is given, as well as theoretical bounds and practical evidence of its feasibility. State of the art results on various benchmark datasets prove the effectiveness of our approach.

Original languageEnglish
Title of host publicationProceedings of the 12th SIAM International Conference on Data Mining, SDM 2012
Pages975-986
Number of pages12
Publication statusPublished - 2012
Externally publishedYes
Event12th SIAM International Conference on Data Mining, SDM 2012 - Anaheim, CA
Duration: 26 Apr 201228 Apr 2012

Other

Other12th SIAM International Conference on Data Mining, SDM 2012
CityAnaheim, CA
Period26/4/1228/4/12

ASJC Scopus subject areas

  • Computer Science Applications

Cite this

Martino, G., Navarin, N., & Sperduti, A. (2012). A tree-based kernel for graphs. In Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012 (pp. 975-986)

A tree-based kernel for graphs. / Martino, Giovanni; Navarin, Nicolò; Sperduti, Alessandro.

Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. 2012. p. 975-986.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Martino, G, Navarin, N & Sperduti, A 2012, A tree-based kernel for graphs. in Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. pp. 975-986, 12th SIAM International Conference on Data Mining, SDM 2012, Anaheim, CA, 26/4/12.
Martino G, Navarin N, Sperduti A. A tree-based kernel for graphs. In Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. 2012. p. 975-986
Martino, Giovanni ; Navarin, Nicolò ; Sperduti, Alessandro. / A tree-based kernel for graphs. Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. 2012. pp. 975-986
@inproceedings{28ce80fed6f442cab08043f484ee129f,
title = "A tree-based kernel for graphs",
abstract = "This paper proposes a new tree-based kernel for graphs. Graphs are decomposed into multisets of ordered Directed Acyclic Graphs (DAGs) and a family of kernels computed by application of tree kernels extended to the DAG domain. We focus our attention on the efficient development of one member of this family. A technique for speeding up the computation is given, as well as theoretical bounds and practical evidence of its feasibility. State of the art results on various benchmark datasets prove the effectiveness of our approach.",
author = "Giovanni Martino and Nicol{\`o} Navarin and Alessandro Sperduti",
year = "2012",
language = "English",
isbn = "9781611972320",
pages = "975--986",
booktitle = "Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012",

}

TY - GEN

T1 - A tree-based kernel for graphs

AU - Martino, Giovanni

AU - Navarin, Nicolò

AU - Sperduti, Alessandro

PY - 2012

Y1 - 2012

N2 - This paper proposes a new tree-based kernel for graphs. Graphs are decomposed into multisets of ordered Directed Acyclic Graphs (DAGs) and a family of kernels computed by application of tree kernels extended to the DAG domain. We focus our attention on the efficient development of one member of this family. A technique for speeding up the computation is given, as well as theoretical bounds and practical evidence of its feasibility. State of the art results on various benchmark datasets prove the effectiveness of our approach.

AB - This paper proposes a new tree-based kernel for graphs. Graphs are decomposed into multisets of ordered Directed Acyclic Graphs (DAGs) and a family of kernels computed by application of tree kernels extended to the DAG domain. We focus our attention on the efficient development of one member of this family. A technique for speeding up the computation is given, as well as theoretical bounds and practical evidence of its feasibility. State of the art results on various benchmark datasets prove the effectiveness of our approach.

UR - http://www.scopus.com/inward/record.url?scp=84880225375&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84880225375&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781611972320

SP - 975

EP - 986

BT - Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012

ER -