By the same authors

A Hypergraph Kernel from Isomorphism Tests

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

Standard

A Hypergraph Kernel from Isomorphism Tests. / Bai, Lu; Ren, Peng; Hancock, Edwin R.

Proceedings of the 22nd International Conference on Pattern Recognition. IEEE Computer Society Press, 2014.

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

Harvard

Bai, L, Ren, P & Hancock, ER 2014, A Hypergraph Kernel from Isomorphism Tests. in Proceedings of the 22nd International Conference on Pattern Recognition. IEEE Computer Society Press, 22nd International Conference on Pattern Recognition, Stockholm, Sweden, 24/08/14.

APA

Bai, L., Ren, P., & Hancock, E. R. (Accepted/In press). A Hypergraph Kernel from Isomorphism Tests. In Proceedings of the 22nd International Conference on Pattern Recognition IEEE Computer Society Press.

Vancouver

Bai L, Ren P, Hancock ER. A Hypergraph Kernel from Isomorphism Tests. In Proceedings of the 22nd International Conference on Pattern Recognition. IEEE Computer Society Press. 2014

Author

Bai, Lu ; Ren, Peng ; Hancock, Edwin R. / A Hypergraph Kernel from Isomorphism Tests. Proceedings of the 22nd International Conference on Pattern Recognition. IEEE Computer Society Press, 2014.

Bibtex - Download

@inproceedings{c5cad9317df246a2a812427ca225327f,
title = "A Hypergraph Kernel from Isomorphism Tests",
abstract = "In this paper, we present a hypergraph kernel computed using substructure isomorphism tests. Measuring the isomorphisms between hypergraphs straightforwardly tends to be elusive since a hypergraph may exhibit varying relational orders. We thus transform a hypergraph into a directed line graph. This not only accurately reflects the multiple relationships exhibited by the hypergraph but is also easier to manipulate isomorphism tests. To locate the isomorphisms between hypergraphs through their directed line graphs, we propose a new directed Weisfeiler-Lehman isomorphism test for directed graphs. The new isomorphism test precisely reflects the structure of the directed edges. By identifying the isomorphic substructures of directed graphs, the hypergraph kernel for a pair of hypergraphs is computed by counting the number of pairwise isomorphic substructures from their directed line graphs. We show that our kernel limits tottering that arises in the existing walk and subtree based (hyper)graph kernels. Experiments on challenging (hyper)graph datasets demonstrate the effectiveness of our kernel.",
keywords = "Support vector machines and kernel methods, Statistical, syntactic and structural pattern recognition, Machine learning and data mining",
author = "Lu Bai and Peng Ren and Hancock, {Edwin R.}",
year = "2014",
language = "English",
booktitle = "Proceedings of the 22nd International Conference on Pattern Recognition",
publisher = "IEEE Computer Society Press",
note = "22nd International Conference on Pattern Recognition ; Conference date: 24-08-2014 Through 28-08-2014",

}

RIS (suitable for import to EndNote) - Download

TY - GEN

T1 - A Hypergraph Kernel from Isomorphism Tests

AU - Bai, Lu

AU - Ren, Peng

AU - Hancock, Edwin R.

PY - 2014

Y1 - 2014

N2 - In this paper, we present a hypergraph kernel computed using substructure isomorphism tests. Measuring the isomorphisms between hypergraphs straightforwardly tends to be elusive since a hypergraph may exhibit varying relational orders. We thus transform a hypergraph into a directed line graph. This not only accurately reflects the multiple relationships exhibited by the hypergraph but is also easier to manipulate isomorphism tests. To locate the isomorphisms between hypergraphs through their directed line graphs, we propose a new directed Weisfeiler-Lehman isomorphism test for directed graphs. The new isomorphism test precisely reflects the structure of the directed edges. By identifying the isomorphic substructures of directed graphs, the hypergraph kernel for a pair of hypergraphs is computed by counting the number of pairwise isomorphic substructures from their directed line graphs. We show that our kernel limits tottering that arises in the existing walk and subtree based (hyper)graph kernels. Experiments on challenging (hyper)graph datasets demonstrate the effectiveness of our kernel.

AB - In this paper, we present a hypergraph kernel computed using substructure isomorphism tests. Measuring the isomorphisms between hypergraphs straightforwardly tends to be elusive since a hypergraph may exhibit varying relational orders. We thus transform a hypergraph into a directed line graph. This not only accurately reflects the multiple relationships exhibited by the hypergraph but is also easier to manipulate isomorphism tests. To locate the isomorphisms between hypergraphs through their directed line graphs, we propose a new directed Weisfeiler-Lehman isomorphism test for directed graphs. The new isomorphism test precisely reflects the structure of the directed edges. By identifying the isomorphic substructures of directed graphs, the hypergraph kernel for a pair of hypergraphs is computed by counting the number of pairwise isomorphic substructures from their directed line graphs. We show that our kernel limits tottering that arises in the existing walk and subtree based (hyper)graph kernels. Experiments on challenging (hyper)graph datasets demonstrate the effectiveness of our kernel.

KW - Support vector machines and kernel methods

KW - Statistical, syntactic and structural pattern recognition

KW - Machine learning and data mining

M3 - Conference contribution

BT - Proceedings of the 22nd International Conference on Pattern Recognition

PB - IEEE Computer Society Press

T2 - 22nd International Conference on Pattern Recognition

Y2 - 24 August 2014 through 28 August 2014

ER -

York staff: edit these data