HUBify: Efficient Estimation of Central Entities Across Multiplex Layer Compositions

Abhishek Santra, Sanjukta Bhowmick, Sharma Chakravarthy

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

2 Citations (Scopus)

Abstract

Graphs or networks are a natural way to analyze inter-related set of entities. When these entities are associated with a diverse number of features, each denoting a specific perspective, then the representation can be simplified by forming a network of layers (one for each feature) or multiplexes. Vertices with high centrality values in the multiplexes represent the most influential vertices. However, detecting central entities in multiplexes for different combinations of features becomes computationally expensive, as the number of layers increases. In this paper, we address the task of efficiently identifying high centrality vertices for any conjunctive (AND) combination of features (as represented by multiplex layers.) We propose efficient heuristics that only use results from individual layers to identify high degree and high closeness centrality vertices. Our approaches, when applied to real-world, multi-featured datasets such as IMDb and traffic accidents, show that we can identify the high centrality vertices with an average accuracy of more than 70-80% while reducing the overall computational time by at least 30%.

Original languageEnglish (US)
Title of host publicationProceeding - 17th IEEE International Conference on Data Mining Workshops, ICDMW 2017
EditorsRaju Gottumukkala, George Karypis, Vijay Raghavan, Xindong Wu, Lucio Miele, Srinivas Aluru, Xia Ning, Guozhu Dong
PublisherIEEE Computer Society
Pages142-149
Number of pages8
ISBN (Electronic)9781538614808
DOIs
StatePublished - Dec 15 2017
Event17th IEEE International Conference on Data Mining Workshops, ICDMW 2017 - New Orleans, United States
Duration: Nov 18 2017Nov 21 2017

Publication series

NameIEEE International Conference on Data Mining Workshops, ICDMW
Volume2017-November
ISSN (Print)2375-9232
ISSN (Electronic)2375-9259

Other

Other17th IEEE International Conference on Data Mining Workshops, ICDMW 2017
CountryUnited States
CityNew Orleans
Period11/18/1711/21/17

Fingerprint

Highway accidents
Chemical analysis

Keywords

  • Closeness Centrality
  • Degree Centrality
  • Graph Analysis
  • Lossless Composability
  • Multiplexes

ASJC Scopus subject areas

  • Computer Science Applications
  • Software

Cite this

Santra, A., Bhowmick, S., & Chakravarthy, S. (2017). HUBify: Efficient Estimation of Central Entities Across Multiplex Layer Compositions. In R. Gottumukkala, G. Karypis, V. Raghavan, X. Wu, L. Miele, S. Aluru, X. Ning, ... G. Dong (Eds.), Proceeding - 17th IEEE International Conference on Data Mining Workshops, ICDMW 2017 (pp. 142-149). (IEEE International Conference on Data Mining Workshops, ICDMW; Vol. 2017-November). IEEE Computer Society. https://doi.org/10.1109/ICDMW.2017.24

HUBify : Efficient Estimation of Central Entities Across Multiplex Layer Compositions. / Santra, Abhishek; Bhowmick, Sanjukta; Chakravarthy, Sharma.

Proceeding - 17th IEEE International Conference on Data Mining Workshops, ICDMW 2017. ed. / Raju Gottumukkala; George Karypis; Vijay Raghavan; Xindong Wu; Lucio Miele; Srinivas Aluru; Xia Ning; Guozhu Dong. IEEE Computer Society, 2017. p. 142-149 (IEEE International Conference on Data Mining Workshops, ICDMW; Vol. 2017-November).

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

Santra, A, Bhowmick, S & Chakravarthy, S 2017, HUBify: Efficient Estimation of Central Entities Across Multiplex Layer Compositions. in R Gottumukkala, G Karypis, V Raghavan, X Wu, L Miele, S Aluru, X Ning & G Dong (eds), Proceeding - 17th IEEE International Conference on Data Mining Workshops, ICDMW 2017. IEEE International Conference on Data Mining Workshops, ICDMW, vol. 2017-November, IEEE Computer Society, pp. 142-149, 17th IEEE International Conference on Data Mining Workshops, ICDMW 2017, New Orleans, United States, 11/18/17. https://doi.org/10.1109/ICDMW.2017.24
Santra A, Bhowmick S, Chakravarthy S. HUBify: Efficient Estimation of Central Entities Across Multiplex Layer Compositions. In Gottumukkala R, Karypis G, Raghavan V, Wu X, Miele L, Aluru S, Ning X, Dong G, editors, Proceeding - 17th IEEE International Conference on Data Mining Workshops, ICDMW 2017. IEEE Computer Society. 2017. p. 142-149. (IEEE International Conference on Data Mining Workshops, ICDMW). https://doi.org/10.1109/ICDMW.2017.24
Santra, Abhishek ; Bhowmick, Sanjukta ; Chakravarthy, Sharma. / HUBify : Efficient Estimation of Central Entities Across Multiplex Layer Compositions. Proceeding - 17th IEEE International Conference on Data Mining Workshops, ICDMW 2017. editor / Raju Gottumukkala ; George Karypis ; Vijay Raghavan ; Xindong Wu ; Lucio Miele ; Srinivas Aluru ; Xia Ning ; Guozhu Dong. IEEE Computer Society, 2017. pp. 142-149 (IEEE International Conference on Data Mining Workshops, ICDMW).
@inproceedings{3e11b0a4c9274b38b178e75cfe2123b8,
title = "HUBify: Efficient Estimation of Central Entities Across Multiplex Layer Compositions",
abstract = "Graphs or networks are a natural way to analyze inter-related set of entities. When these entities are associated with a diverse number of features, each denoting a specific perspective, then the representation can be simplified by forming a network of layers (one for each feature) or multiplexes. Vertices with high centrality values in the multiplexes represent the most influential vertices. However, detecting central entities in multiplexes for different combinations of features becomes computationally expensive, as the number of layers increases. In this paper, we address the task of efficiently identifying high centrality vertices for any conjunctive (AND) combination of features (as represented by multiplex layers.) We propose efficient heuristics that only use results from individual layers to identify high degree and high closeness centrality vertices. Our approaches, when applied to real-world, multi-featured datasets such as IMDb and traffic accidents, show that we can identify the high centrality vertices with an average accuracy of more than 70-80{\%} while reducing the overall computational time by at least 30{\%}.",
keywords = "Closeness Centrality, Degree Centrality, Graph Analysis, Lossless Composability, Multiplexes",
author = "Abhishek Santra and Sanjukta Bhowmick and Sharma Chakravarthy",
year = "2017",
month = "12",
day = "15",
doi = "10.1109/ICDMW.2017.24",
language = "English (US)",
series = "IEEE International Conference on Data Mining Workshops, ICDMW",
publisher = "IEEE Computer Society",
pages = "142--149",
editor = "Raju Gottumukkala and George Karypis and Vijay Raghavan and Xindong Wu and Lucio Miele and Srinivas Aluru and Xia Ning and Guozhu Dong",
booktitle = "Proceeding - 17th IEEE International Conference on Data Mining Workshops, ICDMW 2017",

}

TY - GEN

T1 - HUBify

T2 - Efficient Estimation of Central Entities Across Multiplex Layer Compositions

AU - Santra, Abhishek

AU - Bhowmick, Sanjukta

AU - Chakravarthy, Sharma

PY - 2017/12/15

Y1 - 2017/12/15

N2 - Graphs or networks are a natural way to analyze inter-related set of entities. When these entities are associated with a diverse number of features, each denoting a specific perspective, then the representation can be simplified by forming a network of layers (one for each feature) or multiplexes. Vertices with high centrality values in the multiplexes represent the most influential vertices. However, detecting central entities in multiplexes for different combinations of features becomes computationally expensive, as the number of layers increases. In this paper, we address the task of efficiently identifying high centrality vertices for any conjunctive (AND) combination of features (as represented by multiplex layers.) We propose efficient heuristics that only use results from individual layers to identify high degree and high closeness centrality vertices. Our approaches, when applied to real-world, multi-featured datasets such as IMDb and traffic accidents, show that we can identify the high centrality vertices with an average accuracy of more than 70-80% while reducing the overall computational time by at least 30%.

AB - Graphs or networks are a natural way to analyze inter-related set of entities. When these entities are associated with a diverse number of features, each denoting a specific perspective, then the representation can be simplified by forming a network of layers (one for each feature) or multiplexes. Vertices with high centrality values in the multiplexes represent the most influential vertices. However, detecting central entities in multiplexes for different combinations of features becomes computationally expensive, as the number of layers increases. In this paper, we address the task of efficiently identifying high centrality vertices for any conjunctive (AND) combination of features (as represented by multiplex layers.) We propose efficient heuristics that only use results from individual layers to identify high degree and high closeness centrality vertices. Our approaches, when applied to real-world, multi-featured datasets such as IMDb and traffic accidents, show that we can identify the high centrality vertices with an average accuracy of more than 70-80% while reducing the overall computational time by at least 30%.

KW - Closeness Centrality

KW - Degree Centrality

KW - Graph Analysis

KW - Lossless Composability

KW - Multiplexes

UR - http://www.scopus.com/inward/record.url?scp=85038209133&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85038209133&partnerID=8YFLogxK

U2 - 10.1109/ICDMW.2017.24

DO - 10.1109/ICDMW.2017.24

M3 - Conference contribution

AN - SCOPUS:85038209133

T3 - IEEE International Conference on Data Mining Workshops, ICDMW

SP - 142

EP - 149

BT - Proceeding - 17th IEEE International Conference on Data Mining Workshops, ICDMW 2017

A2 - Gottumukkala, Raju

A2 - Karypis, George

A2 - Raghavan, Vijay

A2 - Wu, Xindong

A2 - Miele, Lucio

A2 - Aluru, Srinivas

A2 - Ning, Xia

A2 - Dong, Guozhu

PB - IEEE Computer Society

ER -