Diversified ranking on large graphs

An optimization viewpoint

Hanghang Tong, Jingrui He, Zhen Wen, Ravi Konuru, Ching Yung Lin

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

51 Citations (Scopus)

Abstract

Diversified ranking on graphs is a fundamental mining task and has a variety of high-impact applications. There are two important open questions here. The first challenge is the measure - how to quantify the goodness of a given top-k ranking list that captures both the relevance and the diversity? The second challenge lies in the algorithmic aspect - how to find an optimal, or near-optimal, top-k ranking list that maximizes the measure we defined in a scalable way? In this paper, we address these challenges from an optimization point of view. Firstly, we propose a goodness measure for a given top-k ranking list. The proposed goodness measure intuitively captures both (a) the relevance between each individual node in the ranking list and the query; and (b) the diversity among different nodes in the ranking list. Moreover, we propose a scalable algorithm (linear wrt the size of the graph) that generates a provably near-optimal solution. The experimental evaluations on real graphs demonstrate its effectiveness and efficiency.

Original languageEnglish (US)
Title of host publicationProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Pages1028-1036
Number of pages9
DOIs
StatePublished - 2011
Externally publishedYes
Event17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD'11 - San Diego, CA, United States
Duration: Aug 21 2011Aug 24 2011

Other

Other17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD'11
CountryUnited States
CitySan Diego, CA
Period8/21/118/24/11

Keywords

  • Diversity
  • Graph mining
  • Ranking
  • Scalability

ASJC Scopus subject areas

  • Software
  • Information Systems

Cite this

Tong, H., He, J., Wen, Z., Konuru, R., & Lin, C. Y. (2011). Diversified ranking on large graphs: An optimization viewpoint. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1028-1036) https://doi.org/10.1145/2020408.2020573

Diversified ranking on large graphs : An optimization viewpoint. / Tong, Hanghang; He, Jingrui; Wen, Zhen; Konuru, Ravi; Lin, Ching Yung.

Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2011. p. 1028-1036.

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

Tong, H, He, J, Wen, Z, Konuru, R & Lin, CY 2011, Diversified ranking on large graphs: An optimization viewpoint. in Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. pp. 1028-1036, 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD'11, San Diego, CA, United States, 8/21/11. https://doi.org/10.1145/2020408.2020573
Tong H, He J, Wen Z, Konuru R, Lin CY. Diversified ranking on large graphs: An optimization viewpoint. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2011. p. 1028-1036 https://doi.org/10.1145/2020408.2020573
Tong, Hanghang ; He, Jingrui ; Wen, Zhen ; Konuru, Ravi ; Lin, Ching Yung. / Diversified ranking on large graphs : An optimization viewpoint. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2011. pp. 1028-1036
@inproceedings{c7f003225d2148178868b7c12d8035ca,
title = "Diversified ranking on large graphs: An optimization viewpoint",
abstract = "Diversified ranking on graphs is a fundamental mining task and has a variety of high-impact applications. There are two important open questions here. The first challenge is the measure - how to quantify the goodness of a given top-k ranking list that captures both the relevance and the diversity? The second challenge lies in the algorithmic aspect - how to find an optimal, or near-optimal, top-k ranking list that maximizes the measure we defined in a scalable way? In this paper, we address these challenges from an optimization point of view. Firstly, we propose a goodness measure for a given top-k ranking list. The proposed goodness measure intuitively captures both (a) the relevance between each individual node in the ranking list and the query; and (b) the diversity among different nodes in the ranking list. Moreover, we propose a scalable algorithm (linear wrt the size of the graph) that generates a provably near-optimal solution. The experimental evaluations on real graphs demonstrate its effectiveness and efficiency.",
keywords = "Diversity, Graph mining, Ranking, Scalability",
author = "Hanghang Tong and Jingrui He and Zhen Wen and Ravi Konuru and Lin, {Ching Yung}",
year = "2011",
doi = "10.1145/2020408.2020573",
language = "English (US)",
isbn = "9781450308137",
pages = "1028--1036",
booktitle = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

}

TY - GEN

T1 - Diversified ranking on large graphs

T2 - An optimization viewpoint

AU - Tong, Hanghang

AU - He, Jingrui

AU - Wen, Zhen

AU - Konuru, Ravi

AU - Lin, Ching Yung

PY - 2011

Y1 - 2011

N2 - Diversified ranking on graphs is a fundamental mining task and has a variety of high-impact applications. There are two important open questions here. The first challenge is the measure - how to quantify the goodness of a given top-k ranking list that captures both the relevance and the diversity? The second challenge lies in the algorithmic aspect - how to find an optimal, or near-optimal, top-k ranking list that maximizes the measure we defined in a scalable way? In this paper, we address these challenges from an optimization point of view. Firstly, we propose a goodness measure for a given top-k ranking list. The proposed goodness measure intuitively captures both (a) the relevance between each individual node in the ranking list and the query; and (b) the diversity among different nodes in the ranking list. Moreover, we propose a scalable algorithm (linear wrt the size of the graph) that generates a provably near-optimal solution. The experimental evaluations on real graphs demonstrate its effectiveness and efficiency.

AB - Diversified ranking on graphs is a fundamental mining task and has a variety of high-impact applications. There are two important open questions here. The first challenge is the measure - how to quantify the goodness of a given top-k ranking list that captures both the relevance and the diversity? The second challenge lies in the algorithmic aspect - how to find an optimal, or near-optimal, top-k ranking list that maximizes the measure we defined in a scalable way? In this paper, we address these challenges from an optimization point of view. Firstly, we propose a goodness measure for a given top-k ranking list. The proposed goodness measure intuitively captures both (a) the relevance between each individual node in the ranking list and the query; and (b) the diversity among different nodes in the ranking list. Moreover, we propose a scalable algorithm (linear wrt the size of the graph) that generates a provably near-optimal solution. The experimental evaluations on real graphs demonstrate its effectiveness and efficiency.

KW - Diversity

KW - Graph mining

KW - Ranking

KW - Scalability

UR - http://www.scopus.com/inward/record.url?scp=80052660788&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052660788&partnerID=8YFLogxK

U2 - 10.1145/2020408.2020573

DO - 10.1145/2020408.2020573

M3 - Conference contribution

SN - 9781450308137

SP - 1028

EP - 1036

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

ER -