Opinion retrieval for twitter using extrinsic information

Yoon Sung Kim, Young In Song, Hae-Chang Rim

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Opinion retrieval in social networks is a very useful field for industry because it can provide a facility for monitoring opinions about a product, person or issue in real time. An opinion retrieval system generally retrieves topically relevant and subjective documents based on topical relevance and a degree of subjectivity. Previous studies on opinion retrieval only considered the intrinsic features of original tweet documents and thus suffer from the data sparseness problem. In this paper, we propose a method of utilizing the extrinsic information of the original tweet and solving the data sparseness problem. We have found useful extrinsic features of related tweets, which can properly measure the degree of subjectivity of the original tweet. When we performed an opinion retrieval experiment including proposed extrinsic features within a learning-to-rank framework, the proposed model significantly outperformed both the baseline system and the state-of-the-art opinion retrieval system in terms of Mean Average Precision (MAP) and Precision@K (P@K) metrics.

Original languageEnglish
Pages (from-to)608-629
Number of pages22
JournalJournal of Universal Computer Science
Volume22
Issue number5
Publication statusPublished - 2016

Fingerprint

Retrieval
Monitoring
Industry
Experiments
Social Networks
Baseline
Person
Metric
Experiment
Model

Keywords

  • Opinion Mining
  • Opinion Retrieval
  • Sentiment Analysis
  • Social Media

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Opinion retrieval for twitter using extrinsic information. / Kim, Yoon Sung; Song, Young In; Rim, Hae-Chang.

In: Journal of Universal Computer Science, Vol. 22, No. 5, 2016, p. 608-629.

Research output: Contribution to journalArticle

Kim, Yoon Sung ; Song, Young In ; Rim, Hae-Chang. / Opinion retrieval for twitter using extrinsic information. In: Journal of Universal Computer Science. 2016 ; Vol. 22, No. 5. pp. 608-629.
@article{9ce2a42b559147ebbd6fa648240dd41e,
title = "Opinion retrieval for twitter using extrinsic information",
abstract = "Opinion retrieval in social networks is a very useful field for industry because it can provide a facility for monitoring opinions about a product, person or issue in real time. An opinion retrieval system generally retrieves topically relevant and subjective documents based on topical relevance and a degree of subjectivity. Previous studies on opinion retrieval only considered the intrinsic features of original tweet documents and thus suffer from the data sparseness problem. In this paper, we propose a method of utilizing the extrinsic information of the original tweet and solving the data sparseness problem. We have found useful extrinsic features of related tweets, which can properly measure the degree of subjectivity of the original tweet. When we performed an opinion retrieval experiment including proposed extrinsic features within a learning-to-rank framework, the proposed model significantly outperformed both the baseline system and the state-of-the-art opinion retrieval system in terms of Mean Average Precision (MAP) and Precision@K (P@K) metrics.",
keywords = "Opinion Mining, Opinion Retrieval, Sentiment Analysis, Social Media",
author = "Kim, {Yoon Sung} and Song, {Young In} and Hae-Chang Rim",
year = "2016",
language = "English",
volume = "22",
pages = "608--629",
journal = "Journal of Universal Computer Science",
issn = "0948-6968",
publisher = "Springer Verlag",
number = "5",

}

TY - JOUR

T1 - Opinion retrieval for twitter using extrinsic information

AU - Kim, Yoon Sung

AU - Song, Young In

AU - Rim, Hae-Chang

PY - 2016

Y1 - 2016

N2 - Opinion retrieval in social networks is a very useful field for industry because it can provide a facility for monitoring opinions about a product, person or issue in real time. An opinion retrieval system generally retrieves topically relevant and subjective documents based on topical relevance and a degree of subjectivity. Previous studies on opinion retrieval only considered the intrinsic features of original tweet documents and thus suffer from the data sparseness problem. In this paper, we propose a method of utilizing the extrinsic information of the original tweet and solving the data sparseness problem. We have found useful extrinsic features of related tweets, which can properly measure the degree of subjectivity of the original tweet. When we performed an opinion retrieval experiment including proposed extrinsic features within a learning-to-rank framework, the proposed model significantly outperformed both the baseline system and the state-of-the-art opinion retrieval system in terms of Mean Average Precision (MAP) and Precision@K (P@K) metrics.

AB - Opinion retrieval in social networks is a very useful field for industry because it can provide a facility for monitoring opinions about a product, person or issue in real time. An opinion retrieval system generally retrieves topically relevant and subjective documents based on topical relevance and a degree of subjectivity. Previous studies on opinion retrieval only considered the intrinsic features of original tweet documents and thus suffer from the data sparseness problem. In this paper, we propose a method of utilizing the extrinsic information of the original tweet and solving the data sparseness problem. We have found useful extrinsic features of related tweets, which can properly measure the degree of subjectivity of the original tweet. When we performed an opinion retrieval experiment including proposed extrinsic features within a learning-to-rank framework, the proposed model significantly outperformed both the baseline system and the state-of-the-art opinion retrieval system in terms of Mean Average Precision (MAP) and Precision@K (P@K) metrics.

KW - Opinion Mining

KW - Opinion Retrieval

KW - Sentiment Analysis

KW - Social Media

UR - http://www.scopus.com/inward/record.url?scp=84982948264&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84982948264&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84982948264

VL - 22

SP - 608

EP - 629

JO - Journal of Universal Computer Science

JF - Journal of Universal Computer Science

SN - 0948-6968

IS - 5

ER -