Tweet ranking based on heterogeneous networks

Hongzhao Huang, Arkaitz Zubiaga, Heng Ji, Hongbo Deng, Dong Wang, Hieu Le, Tarek Abdelzaher, Jiawei Han, Alice Leung, John Hancock, Clare Voss

Research output: Contribution to conferencePaper

Abstract

Ranking tweets is a fundamental task to make it easier to distill the vast amounts of information shared by users. In this paper, we explore the novel idea of ranking tweets on a topic using heterogeneous networks. We construct heterogeneous networks by harnessing cross-genre linkages between tweets and semantically-related web documents from formal genres, and inferring implicit links between tweets and users. To rank tweets effectively by capturing the semantics and importance of different linkages, we introduce Tri-HITS, a model to iteratively propagate ranking scores across heterogeneous networks. We show that integrating both formal genre and inferred social networks with tweet networks produces a higher-quality ranking than the tweet networks alone.

Original languageEnglish (US)
Pages1239-1256
Number of pages18
StatePublished - Dec 1 2012
Event24th International Conference on Computational Linguistics, COLING 2012 - Mumbai, India
Duration: Dec 8 2012Dec 15 2012

Other

Other24th International Conference on Computational Linguistics, COLING 2012
CountryIndia
CityMumbai
Period12/8/1212/15/12

Fingerprint

Heterogeneous networks
ranking
genre
Semantics
social network
semantics
Ranking
Linkage

Keywords

  • Heterogeneous networks
  • Iterative propagation model
  • Tweet ranking

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Language and Linguistics
  • Linguistics and Language

Cite this

Huang, H., Zubiaga, A., Ji, H., Deng, H., Wang, D., Le, H., ... Voss, C. (2012). Tweet ranking based on heterogeneous networks. 1239-1256. Paper presented at 24th International Conference on Computational Linguistics, COLING 2012, Mumbai, India.

Tweet ranking based on heterogeneous networks. / Huang, Hongzhao; Zubiaga, Arkaitz; Ji, Heng; Deng, Hongbo; Wang, Dong; Le, Hieu; Abdelzaher, Tarek; Han, Jiawei; Leung, Alice; Hancock, John; Voss, Clare.

2012. 1239-1256 Paper presented at 24th International Conference on Computational Linguistics, COLING 2012, Mumbai, India.

Research output: Contribution to conferencePaper

Huang, H, Zubiaga, A, Ji, H, Deng, H, Wang, D, Le, H, Abdelzaher, T, Han, J, Leung, A, Hancock, J & Voss, C 2012, 'Tweet ranking based on heterogeneous networks', Paper presented at 24th International Conference on Computational Linguistics, COLING 2012, Mumbai, India, 12/8/12 - 12/15/12 pp. 1239-1256.
Huang H, Zubiaga A, Ji H, Deng H, Wang D, Le H et al. Tweet ranking based on heterogeneous networks. 2012. Paper presented at 24th International Conference on Computational Linguistics, COLING 2012, Mumbai, India.
Huang, Hongzhao ; Zubiaga, Arkaitz ; Ji, Heng ; Deng, Hongbo ; Wang, Dong ; Le, Hieu ; Abdelzaher, Tarek ; Han, Jiawei ; Leung, Alice ; Hancock, John ; Voss, Clare. / Tweet ranking based on heterogeneous networks. Paper presented at 24th International Conference on Computational Linguistics, COLING 2012, Mumbai, India.18 p.
@conference{5c10bd4d9e8b447e8154da8a8ceee9cc,
title = "Tweet ranking based on heterogeneous networks",
abstract = "Ranking tweets is a fundamental task to make it easier to distill the vast amounts of information shared by users. In this paper, we explore the novel idea of ranking tweets on a topic using heterogeneous networks. We construct heterogeneous networks by harnessing cross-genre linkages between tweets and semantically-related web documents from formal genres, and inferring implicit links between tweets and users. To rank tweets effectively by capturing the semantics and importance of different linkages, we introduce Tri-HITS, a model to iteratively propagate ranking scores across heterogeneous networks. We show that integrating both formal genre and inferred social networks with tweet networks produces a higher-quality ranking than the tweet networks alone.",
keywords = "Heterogeneous networks, Iterative propagation model, Tweet ranking",
author = "Hongzhao Huang and Arkaitz Zubiaga and Heng Ji and Hongbo Deng and Dong Wang and Hieu Le and Tarek Abdelzaher and Jiawei Han and Alice Leung and John Hancock and Clare Voss",
year = "2012",
month = "12",
day = "1",
language = "English (US)",
pages = "1239--1256",
note = "24th International Conference on Computational Linguistics, COLING 2012 ; Conference date: 08-12-2012 Through 15-12-2012",

}

TY - CONF

T1 - Tweet ranking based on heterogeneous networks

AU - Huang, Hongzhao

AU - Zubiaga, Arkaitz

AU - Ji, Heng

AU - Deng, Hongbo

AU - Wang, Dong

AU - Le, Hieu

AU - Abdelzaher, Tarek

AU - Han, Jiawei

AU - Leung, Alice

AU - Hancock, John

AU - Voss, Clare

PY - 2012/12/1

Y1 - 2012/12/1

N2 - Ranking tweets is a fundamental task to make it easier to distill the vast amounts of information shared by users. In this paper, we explore the novel idea of ranking tweets on a topic using heterogeneous networks. We construct heterogeneous networks by harnessing cross-genre linkages between tweets and semantically-related web documents from formal genres, and inferring implicit links between tweets and users. To rank tweets effectively by capturing the semantics and importance of different linkages, we introduce Tri-HITS, a model to iteratively propagate ranking scores across heterogeneous networks. We show that integrating both formal genre and inferred social networks with tweet networks produces a higher-quality ranking than the tweet networks alone.

AB - Ranking tweets is a fundamental task to make it easier to distill the vast amounts of information shared by users. In this paper, we explore the novel idea of ranking tweets on a topic using heterogeneous networks. We construct heterogeneous networks by harnessing cross-genre linkages between tweets and semantically-related web documents from formal genres, and inferring implicit links between tweets and users. To rank tweets effectively by capturing the semantics and importance of different linkages, we introduce Tri-HITS, a model to iteratively propagate ranking scores across heterogeneous networks. We show that integrating both formal genre and inferred social networks with tweet networks produces a higher-quality ranking than the tweet networks alone.

KW - Heterogeneous networks

KW - Iterative propagation model

KW - Tweet ranking

UR - http://www.scopus.com/inward/record.url?scp=84876814675&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876814675&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:84876814675

SP - 1239

EP - 1256

ER -