Fast direction-aware proximity for graph mining

Hanghang Tong, Christos Faloutsos, Yehuda Koren

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

56 Citations (Scopus)

Abstract

In this paper we study asymmetric proximity measures on directed graphs, which quantify the relationships between two nodes or two groups of nodes. The measures are useful in several graph mining tasks, including clustering, link prediction and connection subgraph discovery. Our proximity measure is based on the conceptof escape probability. This way, we strive to summarize the multiple facets of nodes-proximity, while avoiding some of the pitfalls to which alternative proximity measures are susceptible. A unique feature of the measures is accounting for the underlying directional information. We put a special emphasis on computational efficiency, and develop fast solutions that are applicable in several settings. Our experimental study shows the usefulness of our proposed direction-aware proximity method for several applications, and that our algorithms achieve a significant speedup (up to 50,000x) over straight forward implementations.

Original languageEnglish (US)
Title of host publicationProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Pages747-756
Number of pages10
DOIs
StatePublished - 2007
Externally publishedYes
EventKDD-2007: 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - San Jose, CA, United States
Duration: Aug 12 2007Aug 15 2007

Other

OtherKDD-2007: 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
CountryUnited States
CitySan Jose, CA
Period8/12/078/15/07

Fingerprint

Directed graphs
Computational efficiency

Keywords

  • Graph mining
  • Proximity
  • Random walk

ASJC Scopus subject areas

  • Information Systems

Cite this

Tong, H., Faloutsos, C., & Koren, Y. (2007). Fast direction-aware proximity for graph mining. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 747-756) https://doi.org/10.1145/1281192.1281272

Fast direction-aware proximity for graph mining. / Tong, Hanghang; Faloutsos, Christos; Koren, Yehuda.

Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2007. p. 747-756.

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

Tong, H, Faloutsos, C & Koren, Y 2007, Fast direction-aware proximity for graph mining. in Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. pp. 747-756, KDD-2007: 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Jose, CA, United States, 8/12/07. https://doi.org/10.1145/1281192.1281272
Tong H, Faloutsos C, Koren Y. Fast direction-aware proximity for graph mining. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2007. p. 747-756 https://doi.org/10.1145/1281192.1281272
Tong, Hanghang ; Faloutsos, Christos ; Koren, Yehuda. / Fast direction-aware proximity for graph mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2007. pp. 747-756
@inproceedings{392ae8da23c7457c838f6aaba38bb088,
title = "Fast direction-aware proximity for graph mining",
abstract = "In this paper we study asymmetric proximity measures on directed graphs, which quantify the relationships between two nodes or two groups of nodes. The measures are useful in several graph mining tasks, including clustering, link prediction and connection subgraph discovery. Our proximity measure is based on the conceptof escape probability. This way, we strive to summarize the multiple facets of nodes-proximity, while avoiding some of the pitfalls to which alternative proximity measures are susceptible. A unique feature of the measures is accounting for the underlying directional information. We put a special emphasis on computational efficiency, and develop fast solutions that are applicable in several settings. Our experimental study shows the usefulness of our proposed direction-aware proximity method for several applications, and that our algorithms achieve a significant speedup (up to 50,000x) over straight forward implementations.",
keywords = "Graph mining, Proximity, Random walk",
author = "Hanghang Tong and Christos Faloutsos and Yehuda Koren",
year = "2007",
doi = "10.1145/1281192.1281272",
language = "English (US)",
isbn = "1595936092",
pages = "747--756",
booktitle = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

}

TY - GEN

T1 - Fast direction-aware proximity for graph mining

AU - Tong, Hanghang

AU - Faloutsos, Christos

AU - Koren, Yehuda

PY - 2007

Y1 - 2007

N2 - In this paper we study asymmetric proximity measures on directed graphs, which quantify the relationships between two nodes or two groups of nodes. The measures are useful in several graph mining tasks, including clustering, link prediction and connection subgraph discovery. Our proximity measure is based on the conceptof escape probability. This way, we strive to summarize the multiple facets of nodes-proximity, while avoiding some of the pitfalls to which alternative proximity measures are susceptible. A unique feature of the measures is accounting for the underlying directional information. We put a special emphasis on computational efficiency, and develop fast solutions that are applicable in several settings. Our experimental study shows the usefulness of our proposed direction-aware proximity method for several applications, and that our algorithms achieve a significant speedup (up to 50,000x) over straight forward implementations.

AB - In this paper we study asymmetric proximity measures on directed graphs, which quantify the relationships between two nodes or two groups of nodes. The measures are useful in several graph mining tasks, including clustering, link prediction and connection subgraph discovery. Our proximity measure is based on the conceptof escape probability. This way, we strive to summarize the multiple facets of nodes-proximity, while avoiding some of the pitfalls to which alternative proximity measures are susceptible. A unique feature of the measures is accounting for the underlying directional information. We put a special emphasis on computational efficiency, and develop fast solutions that are applicable in several settings. Our experimental study shows the usefulness of our proposed direction-aware proximity method for several applications, and that our algorithms achieve a significant speedup (up to 50,000x) over straight forward implementations.

KW - Graph mining

KW - Proximity

KW - Random walk

UR - http://www.scopus.com/inward/record.url?scp=36849093960&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=36849093960&partnerID=8YFLogxK

U2 - 10.1145/1281192.1281272

DO - 10.1145/1281192.1281272

M3 - Conference contribution

SN - 1595936092

SN - 9781595936097

SP - 747

EP - 756

BT - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

ER -