Fast random walk graph kernel

U. Kang, Hanghang Tong, Jimeng Sun

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

37 Citations (Scopus)

Abstract

Random walk graph kernel has been used as an important tool for various data mining tasks including classi fication and similarity computation. Despite its usefulness, however, it suffers from the expensive computational cost which is at least O(n3) or O(m2) for graphs with n nodes and m edges. In this paper, we propose Ark, a set of fast algorithms for random walk graph kernel computation. Ark is based on the observation that real graphs have much lower intrinsic ranks, compared with the orders of the graphs. Ark exploits the low rank structure to quickly compute random walk graph kernels in O(n 2) or O(m) time. Experimental results show that our method is up to 97,865× faster than the existing algorithms, while providing more than 91.3% of the accuracies.

Original languageEnglish (US)
Title of host publicationProceedings of the 12th SIAM International Conference on Data Mining, SDM 2012
Pages828-838
Number of pages11
StatePublished - 2012
Externally publishedYes
Event12th SIAM International Conference on Data Mining, SDM 2012 - Anaheim, CA, United States
Duration: Apr 26 2012Apr 28 2012

Other

Other12th SIAM International Conference on Data Mining, SDM 2012
CountryUnited States
CityAnaheim, CA
Period4/26/124/28/12

Fingerprint

Data mining
Costs

ASJC Scopus subject areas

  • Computer Science Applications

Cite this

Kang, U., Tong, H., & Sun, J. (2012). Fast random walk graph kernel. In Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012 (pp. 828-838)

Fast random walk graph kernel. / Kang, U.; Tong, Hanghang; Sun, Jimeng.

Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. 2012. p. 828-838.

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

Kang, U, Tong, H & Sun, J 2012, Fast random walk graph kernel. in Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. pp. 828-838, 12th SIAM International Conference on Data Mining, SDM 2012, Anaheim, CA, United States, 4/26/12.
Kang U, Tong H, Sun J. Fast random walk graph kernel. In Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. 2012. p. 828-838
Kang, U. ; Tong, Hanghang ; Sun, Jimeng. / Fast random walk graph kernel. Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. 2012. pp. 828-838
@inproceedings{4723c50dc5804042b063eee433fcf52a,
title = "Fast random walk graph kernel",
abstract = "Random walk graph kernel has been used as an important tool for various data mining tasks including classi fication and similarity computation. Despite its usefulness, however, it suffers from the expensive computational cost which is at least O(n3) or O(m2) for graphs with n nodes and m edges. In this paper, we propose Ark, a set of fast algorithms for random walk graph kernel computation. Ark is based on the observation that real graphs have much lower intrinsic ranks, compared with the orders of the graphs. Ark exploits the low rank structure to quickly compute random walk graph kernels in O(n 2) or O(m) time. Experimental results show that our method is up to 97,865× faster than the existing algorithms, while providing more than 91.3{\%} of the accuracies.",
author = "U. Kang and Hanghang Tong and Jimeng Sun",
year = "2012",
language = "English (US)",
isbn = "9781611972320",
pages = "828--838",
booktitle = "Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012",

}

TY - GEN

T1 - Fast random walk graph kernel

AU - Kang, U.

AU - Tong, Hanghang

AU - Sun, Jimeng

PY - 2012

Y1 - 2012

N2 - Random walk graph kernel has been used as an important tool for various data mining tasks including classi fication and similarity computation. Despite its usefulness, however, it suffers from the expensive computational cost which is at least O(n3) or O(m2) for graphs with n nodes and m edges. In this paper, we propose Ark, a set of fast algorithms for random walk graph kernel computation. Ark is based on the observation that real graphs have much lower intrinsic ranks, compared with the orders of the graphs. Ark exploits the low rank structure to quickly compute random walk graph kernels in O(n 2) or O(m) time. Experimental results show that our method is up to 97,865× faster than the existing algorithms, while providing more than 91.3% of the accuracies.

AB - Random walk graph kernel has been used as an important tool for various data mining tasks including classi fication and similarity computation. Despite its usefulness, however, it suffers from the expensive computational cost which is at least O(n3) or O(m2) for graphs with n nodes and m edges. In this paper, we propose Ark, a set of fast algorithms for random walk graph kernel computation. Ark is based on the observation that real graphs have much lower intrinsic ranks, compared with the orders of the graphs. Ark exploits the low rank structure to quickly compute random walk graph kernels in O(n 2) or O(m) time. Experimental results show that our method is up to 97,865× faster than the existing algorithms, while providing more than 91.3% of the accuracies.

UR - http://www.scopus.com/inward/record.url?scp=84880238353&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84880238353&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781611972320

SP - 828

EP - 838

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

ER -