Abstract

Social media has reshaped the way in which people interact with each other. The rapid development of participatory web and social networking sites like YouTube, Twitter, and Facebook, also brings about many data mining opportunities and novel challenges. In particular, we focus on classification tasks with user interaction information in a social network. Networks in social media are heterogeneous, consisting of various relations. Since the relation-type information may not be available in social media, most existing approaches treat these inhomogeneous connections homogeneously, leading to an unsatisfactory classification performance. In order to handle the network heterogeneity, we propose the concept of social dimension to represent actors' latent affiliations, and develop a classification framework based on that. The proposed framework, SocioDim, first extracts social dimensions based on the network structure to accurately capture prominent interaction patterns between actors, then learns a discriminative classifier to select relevant social dimensions. SocioDim, by differentiating different types of network connections, outperforms existing representative methods of classification in social media, and offers a simple yet effective approach to integrating two types of seemingly orthogonal information: the network of actors and their attributes.

Original languageEnglish (US)
Pages (from-to)447-478
Number of pages32
JournalData Mining and Knowledge Discovery
Volume23
Issue number3
DOIs
StatePublished - Nov 2011

Fingerprint

Data mining
Classifiers

Keywords

  • Collective inference
  • Relational learning
  • Social media
  • Social network analysis
  • Within-network classification

ASJC Scopus subject areas

  • Information Systems
  • Computer Science Applications
  • Computer Networks and Communications

Cite this

Leveraging social media networks for classification. / Tang, Lei; Liu, Huan.

In: Data Mining and Knowledge Discovery, Vol. 23, No. 3, 11.2011, p. 447-478.

Research output: Contribution to journalArticle

@article{3b05f4027bae44e4868c8521b837198f,
title = "Leveraging social media networks for classification",
abstract = "Social media has reshaped the way in which people interact with each other. The rapid development of participatory web and social networking sites like YouTube, Twitter, and Facebook, also brings about many data mining opportunities and novel challenges. In particular, we focus on classification tasks with user interaction information in a social network. Networks in social media are heterogeneous, consisting of various relations. Since the relation-type information may not be available in social media, most existing approaches treat these inhomogeneous connections homogeneously, leading to an unsatisfactory classification performance. In order to handle the network heterogeneity, we propose the concept of social dimension to represent actors' latent affiliations, and develop a classification framework based on that. The proposed framework, SocioDim, first extracts social dimensions based on the network structure to accurately capture prominent interaction patterns between actors, then learns a discriminative classifier to select relevant social dimensions. SocioDim, by differentiating different types of network connections, outperforms existing representative methods of classification in social media, and offers a simple yet effective approach to integrating two types of seemingly orthogonal information: the network of actors and their attributes.",
keywords = "Collective inference, Relational learning, Social media, Social network analysis, Within-network classification",
author = "Lei Tang and Huan Liu",
year = "2011",
month = "11",
doi = "10.1007/s10618-010-0210-x",
language = "English (US)",
volume = "23",
pages = "447--478",
journal = "Data Mining and Knowledge Discovery",
issn = "1384-5810",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - Leveraging social media networks for classification

AU - Tang, Lei

AU - Liu, Huan

PY - 2011/11

Y1 - 2011/11

N2 - Social media has reshaped the way in which people interact with each other. The rapid development of participatory web and social networking sites like YouTube, Twitter, and Facebook, also brings about many data mining opportunities and novel challenges. In particular, we focus on classification tasks with user interaction information in a social network. Networks in social media are heterogeneous, consisting of various relations. Since the relation-type information may not be available in social media, most existing approaches treat these inhomogeneous connections homogeneously, leading to an unsatisfactory classification performance. In order to handle the network heterogeneity, we propose the concept of social dimension to represent actors' latent affiliations, and develop a classification framework based on that. The proposed framework, SocioDim, first extracts social dimensions based on the network structure to accurately capture prominent interaction patterns between actors, then learns a discriminative classifier to select relevant social dimensions. SocioDim, by differentiating different types of network connections, outperforms existing representative methods of classification in social media, and offers a simple yet effective approach to integrating two types of seemingly orthogonal information: the network of actors and their attributes.

AB - Social media has reshaped the way in which people interact with each other. The rapid development of participatory web and social networking sites like YouTube, Twitter, and Facebook, also brings about many data mining opportunities and novel challenges. In particular, we focus on classification tasks with user interaction information in a social network. Networks in social media are heterogeneous, consisting of various relations. Since the relation-type information may not be available in social media, most existing approaches treat these inhomogeneous connections homogeneously, leading to an unsatisfactory classification performance. In order to handle the network heterogeneity, we propose the concept of social dimension to represent actors' latent affiliations, and develop a classification framework based on that. The proposed framework, SocioDim, first extracts social dimensions based on the network structure to accurately capture prominent interaction patterns between actors, then learns a discriminative classifier to select relevant social dimensions. SocioDim, by differentiating different types of network connections, outperforms existing representative methods of classification in social media, and offers a simple yet effective approach to integrating two types of seemingly orthogonal information: the network of actors and their attributes.

KW - Collective inference

KW - Relational learning

KW - Social media

KW - Social network analysis

KW - Within-network classification

UR - http://www.scopus.com/inward/record.url?scp=80855134335&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80855134335&partnerID=8YFLogxK

U2 - 10.1007/s10618-010-0210-x

DO - 10.1007/s10618-010-0210-x

M3 - Article

VL - 23

SP - 447

EP - 478

JO - Data Mining and Knowledge Discovery

JF - Data Mining and Knowledge Discovery

SN - 1384-5810

IS - 3

ER -