Abstract

Social network data contains many hidden relationships. The most well known is the communities formed by users. Moreover, typical social network data, such as Twitter, can also be interpreted in terms of three-dimensional relationships; namely the users, issues discussed by the users, and terminology chosen by the users in these discussions. In this paper, we propose a new problem to generate co-clusters in these three dimensions simultaneously. There are three major differences between our problem and the standard co-clustering problem definition: a node can be a member of more than one clusters; all the nodes are not necessarily members of some cluster; and edges are signed and cluster are expected to have high density of positive signed edges, and low density of negative signed edges. We apply our method to the tweets of British politicians just before the Brexit referendum. Our motivation is to discover clusters of politicians, issues and the sentimental words politicians use to express their feelings on these issues in their tweets.

Original languageEnglish (US)
Pages (from-to)79-94
Number of pages16
JournalExpert Systems with Applications
Volume100
DOIs
StatePublished - Jun 15 2018

Fingerprint

Terminology

Keywords

  • 3 partite graph
  • Co-clustering
  • Hypergraph
  • Sentiment analysis
  • Social network analysis

ASJC Scopus subject areas

  • Engineering(all)
  • Computer Science Applications
  • Artificial Intelligence

Cite this

Triadic co-clustering of users, issues and sentiments in political tweets. / Koç, Sefa Şahin; Özer, Mert; Toroslu, İsmail Hakkı; Davulcu, Hasan; Jordan, Jeremy.

In: Expert Systems with Applications, Vol. 100, 15.06.2018, p. 79-94.

Research output: Contribution to journalArticle

Koç, Sefa Şahin ; Özer, Mert ; Toroslu, İsmail Hakkı ; Davulcu, Hasan ; Jordan, Jeremy. / Triadic co-clustering of users, issues and sentiments in political tweets. In: Expert Systems with Applications. 2018 ; Vol. 100. pp. 79-94.
@article{a1a0412bfcb5468495cca493807f9a17,
title = "Triadic co-clustering of users, issues and sentiments in political tweets",
abstract = "Social network data contains many hidden relationships. The most well known is the communities formed by users. Moreover, typical social network data, such as Twitter, can also be interpreted in terms of three-dimensional relationships; namely the users, issues discussed by the users, and terminology chosen by the users in these discussions. In this paper, we propose a new problem to generate co-clusters in these three dimensions simultaneously. There are three major differences between our problem and the standard co-clustering problem definition: a node can be a member of more than one clusters; all the nodes are not necessarily members of some cluster; and edges are signed and cluster are expected to have high density of positive signed edges, and low density of negative signed edges. We apply our method to the tweets of British politicians just before the Brexit referendum. Our motivation is to discover clusters of politicians, issues and the sentimental words politicians use to express their feelings on these issues in their tweets.",
keywords = "3 partite graph, Co-clustering, Hypergraph, Sentiment analysis, Social network analysis",
author = "Ko{\cc}, {Sefa Şahin} and Mert {\"O}zer and Toroslu, {İsmail Hakkı} and Hasan Davulcu and Jeremy Jordan",
year = "2018",
month = "6",
day = "15",
doi = "10.1016/j.eswa.2018.01.043",
language = "English (US)",
volume = "100",
pages = "79--94",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Triadic co-clustering of users, issues and sentiments in political tweets

AU - Koç, Sefa Şahin

AU - Özer, Mert

AU - Toroslu, İsmail Hakkı

AU - Davulcu, Hasan

AU - Jordan, Jeremy

PY - 2018/6/15

Y1 - 2018/6/15

N2 - Social network data contains many hidden relationships. The most well known is the communities formed by users. Moreover, typical social network data, such as Twitter, can also be interpreted in terms of three-dimensional relationships; namely the users, issues discussed by the users, and terminology chosen by the users in these discussions. In this paper, we propose a new problem to generate co-clusters in these three dimensions simultaneously. There are three major differences between our problem and the standard co-clustering problem definition: a node can be a member of more than one clusters; all the nodes are not necessarily members of some cluster; and edges are signed and cluster are expected to have high density of positive signed edges, and low density of negative signed edges. We apply our method to the tweets of British politicians just before the Brexit referendum. Our motivation is to discover clusters of politicians, issues and the sentimental words politicians use to express their feelings on these issues in their tweets.

AB - Social network data contains many hidden relationships. The most well known is the communities formed by users. Moreover, typical social network data, such as Twitter, can also be interpreted in terms of three-dimensional relationships; namely the users, issues discussed by the users, and terminology chosen by the users in these discussions. In this paper, we propose a new problem to generate co-clusters in these three dimensions simultaneously. There are three major differences between our problem and the standard co-clustering problem definition: a node can be a member of more than one clusters; all the nodes are not necessarily members of some cluster; and edges are signed and cluster are expected to have high density of positive signed edges, and low density of negative signed edges. We apply our method to the tweets of British politicians just before the Brexit referendum. Our motivation is to discover clusters of politicians, issues and the sentimental words politicians use to express their feelings on these issues in their tweets.

KW - 3 partite graph

KW - Co-clustering

KW - Hypergraph

KW - Sentiment analysis

KW - Social network analysis

UR - http://www.scopus.com/inward/record.url?scp=85041521295&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041521295&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2018.01.043

DO - 10.1016/j.eswa.2018.01.043

M3 - Article

VL - 100

SP - 79

EP - 94

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

ER -