Fast shortest path distance estimation in large networks

Michalis Potamias, Francesco Bonchi, Carlos Castillo, Aristides Gionis

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

153 Citations (Scopus)

Abstract

In this paper we study approximate landmark-based methods for point-to-point distance estimation in very large networks. These methods involve selecting a subset of nodes as landmarks and computing offline the distances from each node in the graph to those landmarks. At runtime, when the distance between a pair of nodes is needed, it can be estimated quickly by combining the precomputed distances. We prove that selecting the optimal set of landmarks is an NP-hard problem, and thus heuristic solutions need to be employed. We therefore explore theoretical insights to devise a variety of simple methods that scale well in very large networks. The efficiency of the suggested techniques is tested experimentally using five real-world graphs having millions of edges. While theoretical bounds support the claim that random landmarks work well in practice, our extensive experimentation shows that smart landmark selection can yield dramatically more accurate results: for a given target accuracy, our methods require as much as 250 times less space than selecting landmarks at random. In addition, we demonstrate that at a very small accuracy loss our techniques are several orders of magnitude faster than the state-of-the-art exact methods. Finally, we study an application of our methods to the task of social search in large graphs.

Original languageEnglish
Title of host publicationInternational Conference on Information and Knowledge Management, Proceedings
Pages867-876
Number of pages10
DOIs
Publication statusPublished - 1 Dec 2009
Externally publishedYes
EventACM 18th International Conference on Information and Knowledge Management, CIKM 2009 - Hong Kong, China
Duration: 2 Nov 20096 Nov 2009

Other

OtherACM 18th International Conference on Information and Knowledge Management, CIKM 2009
CountryChina
CityHong Kong
Period2/11/096/11/09

Fingerprint

Shortest path
Node
Graph
Experimentation
Heuristics
Social search
NP-hard

Keywords

  • Graphs
  • Landmarks methods
  • Shortest-paths

ASJC Scopus subject areas

  • Business, Management and Accounting(all)
  • Decision Sciences(all)

Cite this

Potamias, M., Bonchi, F., Castillo, C., & Gionis, A. (2009). Fast shortest path distance estimation in large networks. In International Conference on Information and Knowledge Management, Proceedings (pp. 867-876) https://doi.org/10.1145/1645953.1646063

Fast shortest path distance estimation in large networks. / Potamias, Michalis; Bonchi, Francesco; Castillo, Carlos; Gionis, Aristides.

International Conference on Information and Knowledge Management, Proceedings. 2009. p. 867-876.

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

Potamias, M, Bonchi, F, Castillo, C & Gionis, A 2009, Fast shortest path distance estimation in large networks. in International Conference on Information and Knowledge Management, Proceedings. pp. 867-876, ACM 18th International Conference on Information and Knowledge Management, CIKM 2009, Hong Kong, China, 2/11/09. https://doi.org/10.1145/1645953.1646063
Potamias M, Bonchi F, Castillo C, Gionis A. Fast shortest path distance estimation in large networks. In International Conference on Information and Knowledge Management, Proceedings. 2009. p. 867-876 https://doi.org/10.1145/1645953.1646063
Potamias, Michalis ; Bonchi, Francesco ; Castillo, Carlos ; Gionis, Aristides. / Fast shortest path distance estimation in large networks. International Conference on Information and Knowledge Management, Proceedings. 2009. pp. 867-876
@inproceedings{f09ce2b01607479b8966c7b9494e7e2e,
title = "Fast shortest path distance estimation in large networks",
abstract = "In this paper we study approximate landmark-based methods for point-to-point distance estimation in very large networks. These methods involve selecting a subset of nodes as landmarks and computing offline the distances from each node in the graph to those landmarks. At runtime, when the distance between a pair of nodes is needed, it can be estimated quickly by combining the precomputed distances. We prove that selecting the optimal set of landmarks is an NP-hard problem, and thus heuristic solutions need to be employed. We therefore explore theoretical insights to devise a variety of simple methods that scale well in very large networks. The efficiency of the suggested techniques is tested experimentally using five real-world graphs having millions of edges. While theoretical bounds support the claim that random landmarks work well in practice, our extensive experimentation shows that smart landmark selection can yield dramatically more accurate results: for a given target accuracy, our methods require as much as 250 times less space than selecting landmarks at random. In addition, we demonstrate that at a very small accuracy loss our techniques are several orders of magnitude faster than the state-of-the-art exact methods. Finally, we study an application of our methods to the task of social search in large graphs.",
keywords = "Graphs, Landmarks methods, Shortest-paths",
author = "Michalis Potamias and Francesco Bonchi and Carlos Castillo and Aristides Gionis",
year = "2009",
month = "12",
day = "1",
doi = "10.1145/1645953.1646063",
language = "English",
isbn = "9781605585123",
pages = "867--876",
booktitle = "International Conference on Information and Knowledge Management, Proceedings",

}

TY - GEN

T1 - Fast shortest path distance estimation in large networks

AU - Potamias, Michalis

AU - Bonchi, Francesco

AU - Castillo, Carlos

AU - Gionis, Aristides

PY - 2009/12/1

Y1 - 2009/12/1

N2 - In this paper we study approximate landmark-based methods for point-to-point distance estimation in very large networks. These methods involve selecting a subset of nodes as landmarks and computing offline the distances from each node in the graph to those landmarks. At runtime, when the distance between a pair of nodes is needed, it can be estimated quickly by combining the precomputed distances. We prove that selecting the optimal set of landmarks is an NP-hard problem, and thus heuristic solutions need to be employed. We therefore explore theoretical insights to devise a variety of simple methods that scale well in very large networks. The efficiency of the suggested techniques is tested experimentally using five real-world graphs having millions of edges. While theoretical bounds support the claim that random landmarks work well in practice, our extensive experimentation shows that smart landmark selection can yield dramatically more accurate results: for a given target accuracy, our methods require as much as 250 times less space than selecting landmarks at random. In addition, we demonstrate that at a very small accuracy loss our techniques are several orders of magnitude faster than the state-of-the-art exact methods. Finally, we study an application of our methods to the task of social search in large graphs.

AB - In this paper we study approximate landmark-based methods for point-to-point distance estimation in very large networks. These methods involve selecting a subset of nodes as landmarks and computing offline the distances from each node in the graph to those landmarks. At runtime, when the distance between a pair of nodes is needed, it can be estimated quickly by combining the precomputed distances. We prove that selecting the optimal set of landmarks is an NP-hard problem, and thus heuristic solutions need to be employed. We therefore explore theoretical insights to devise a variety of simple methods that scale well in very large networks. The efficiency of the suggested techniques is tested experimentally using five real-world graphs having millions of edges. While theoretical bounds support the claim that random landmarks work well in practice, our extensive experimentation shows that smart landmark selection can yield dramatically more accurate results: for a given target accuracy, our methods require as much as 250 times less space than selecting landmarks at random. In addition, we demonstrate that at a very small accuracy loss our techniques are several orders of magnitude faster than the state-of-the-art exact methods. Finally, we study an application of our methods to the task of social search in large graphs.

KW - Graphs

KW - Landmarks methods

KW - Shortest-paths

UR - http://www.scopus.com/inward/record.url?scp=74549131087&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=74549131087&partnerID=8YFLogxK

U2 - 10.1145/1645953.1646063

DO - 10.1145/1645953.1646063

M3 - Conference contribution

SN - 9781605585123

SP - 867

EP - 876

BT - International Conference on Information and Knowledge Management, Proceedings

ER -