Leader identification in an online health community for cancer survivors

a social network-based classification approach

Kang Zhao, Greta E. Greer, John Yen, Prasenjit Mitra, Kenneth Portier

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Online health communities (OHCs) are an important source of social support for cancer survivors and their informal caregivers. This research attempted to identify leaders in a popular online forum for cancer survivors and caregivers using classification techniques. We first extracted user features from many different perspectives, including contributions, network centralities, and linguistic features. Based on these features, we leveraged the structure of the social network among users and generated new neighborhood-based and cluster-based features. Classification results revealed that these features are discriminative for leader identification. Using these features, we developed a hybrid approach based on an ensemble classifier that performs better than many traditional metrics. This research has implications for understanding and managing OHCs.

Original languageEnglish
Pages (from-to)629-645
Number of pages17
JournalInformation Systems and e-Business Management
Volume13
Issue number4
DOIs
Publication statusPublished - 1 Nov 2015
Externally publishedYes

Fingerprint

Health
Linguistics
Classifiers

Keywords

  • Classification
  • Clustering
  • Leader identification
  • Online health community
  • Social network

ASJC Scopus subject areas

  • Information Systems

Cite this

Leader identification in an online health community for cancer survivors : a social network-based classification approach. / Zhao, Kang; Greer, Greta E.; Yen, John; Mitra, Prasenjit; Portier, Kenneth.

In: Information Systems and e-Business Management, Vol. 13, No. 4, 01.11.2015, p. 629-645.

Research output: Contribution to journalArticle

Zhao, Kang ; Greer, Greta E. ; Yen, John ; Mitra, Prasenjit ; Portier, Kenneth. / Leader identification in an online health community for cancer survivors : a social network-based classification approach. In: Information Systems and e-Business Management. 2015 ; Vol. 13, No. 4. pp. 629-645.
@article{6b416d9d9001466e9ea6de4d1f9db4ba,
title = "Leader identification in an online health community for cancer survivors: a social network-based classification approach",
abstract = "Online health communities (OHCs) are an important source of social support for cancer survivors and their informal caregivers. This research attempted to identify leaders in a popular online forum for cancer survivors and caregivers using classification techniques. We first extracted user features from many different perspectives, including contributions, network centralities, and linguistic features. Based on these features, we leveraged the structure of the social network among users and generated new neighborhood-based and cluster-based features. Classification results revealed that these features are discriminative for leader identification. Using these features, we developed a hybrid approach based on an ensemble classifier that performs better than many traditional metrics. This research has implications for understanding and managing OHCs.",
keywords = "Classification, Clustering, Leader identification, Online health community, Social network",
author = "Kang Zhao and Greer, {Greta E.} and John Yen and Prasenjit Mitra and Kenneth Portier",
year = "2015",
month = "11",
day = "1",
doi = "10.1007/s10257-014-0260-5",
language = "English",
volume = "13",
pages = "629--645",
journal = "Information Systems and e-Business Management",
issn = "1617-9846",
publisher = "Springer Verlag",
number = "4",

}

TY - JOUR

T1 - Leader identification in an online health community for cancer survivors

T2 - a social network-based classification approach

AU - Zhao, Kang

AU - Greer, Greta E.

AU - Yen, John

AU - Mitra, Prasenjit

AU - Portier, Kenneth

PY - 2015/11/1

Y1 - 2015/11/1

N2 - Online health communities (OHCs) are an important source of social support for cancer survivors and their informal caregivers. This research attempted to identify leaders in a popular online forum for cancer survivors and caregivers using classification techniques. We first extracted user features from many different perspectives, including contributions, network centralities, and linguistic features. Based on these features, we leveraged the structure of the social network among users and generated new neighborhood-based and cluster-based features. Classification results revealed that these features are discriminative for leader identification. Using these features, we developed a hybrid approach based on an ensemble classifier that performs better than many traditional metrics. This research has implications for understanding and managing OHCs.

AB - Online health communities (OHCs) are an important source of social support for cancer survivors and their informal caregivers. This research attempted to identify leaders in a popular online forum for cancer survivors and caregivers using classification techniques. We first extracted user features from many different perspectives, including contributions, network centralities, and linguistic features. Based on these features, we leveraged the structure of the social network among users and generated new neighborhood-based and cluster-based features. Classification results revealed that these features are discriminative for leader identification. Using these features, we developed a hybrid approach based on an ensemble classifier that performs better than many traditional metrics. This research has implications for understanding and managing OHCs.

KW - Classification

KW - Clustering

KW - Leader identification

KW - Online health community

KW - Social network

UR - http://www.scopus.com/inward/record.url?scp=84948992373&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84948992373&partnerID=8YFLogxK

U2 - 10.1007/s10257-014-0260-5

DO - 10.1007/s10257-014-0260-5

M3 - Article

VL - 13

SP - 629

EP - 645

JO - Information Systems and e-Business Management

JF - Information Systems and e-Business Management

SN - 1617-9846

IS - 4

ER -