Relational features in fine-grained opinion analysis

Richard Johansson, Alessandro Moschitti

Research output: Contribution to journalArticle

32 Citations (Scopus)

Abstract

Fine-grained opinion analysis methods often make use of linguistic features but typically do not take the interaction between opinions into account. This article describes a set of experiments that demonstrate that relational features, mainly derived from dependency-syntactic and semantic role structures, can significantly improve the performance of automatic systems for a number of fine-grained opinion analysis tasks: marking up opinion expressions, finding opinion holders, and determining the polarities of opinion expressions. These features make it possible to model the way opinions expressed in natural-language discourse interact in a sentence over arbitrary distances. The use of relations requires us to consider multiple opinions simultaneously, which makes the search for the optimal analysis intractable. However, a reranker can be used as a sufficiently accurate and efficient approximation. A number of feature sets and machine learning approaches for the rerankers are evaluated. For the task of opinion expression extraction, the best model shows a 10-point absolute improvement in soft recall on the MPQA corpus over a conventional sequence labeler based on local contextual features, while precision decreases only slightly. Significant improvements are also seen for the extended tasks where holders and polarities are considered: 10 and 7 points in recall, respectively. In addition, the systems outperform previously published results for unlabeled (6 F-measure points) and polarity-labeled (10-15 points) opinion expression extraction. Finally, as an extrinsic evaluation, the extracted MPQA-style opinion expressions are used in practical opinion mining tasks. In all scenarios considered, the machine learning features derived from the opinion expressions lead to statistically significant improvements.

Original languageEnglish
Pages (from-to)473-509
Number of pages37
JournalComputational Linguistics
Volume39
Issue number3
DOIs
Publication statusPublished - 1 Sep 2013
Externally publishedYes

Fingerprint

Learning systems
Syntactics
Linguistics
Semantics
Experiments
Polarity
learning
semantics
scenario
linguistics
Machine Learning
discourse

ASJC Scopus subject areas

  • Computer Science Applications
  • Artificial Intelligence
  • Linguistics and Language
  • Language and Linguistics

Cite this

Relational features in fine-grained opinion analysis. / Johansson, Richard; Moschitti, Alessandro.

In: Computational Linguistics, Vol. 39, No. 3, 01.09.2013, p. 473-509.

Research output: Contribution to journalArticle

@article{cc6a2bb0a5e14349a5b0e77010fa66a1,
title = "Relational features in fine-grained opinion analysis",
abstract = "Fine-grained opinion analysis methods often make use of linguistic features but typically do not take the interaction between opinions into account. This article describes a set of experiments that demonstrate that relational features, mainly derived from dependency-syntactic and semantic role structures, can significantly improve the performance of automatic systems for a number of fine-grained opinion analysis tasks: marking up opinion expressions, finding opinion holders, and determining the polarities of opinion expressions. These features make it possible to model the way opinions expressed in natural-language discourse interact in a sentence over arbitrary distances. The use of relations requires us to consider multiple opinions simultaneously, which makes the search for the optimal analysis intractable. However, a reranker can be used as a sufficiently accurate and efficient approximation. A number of feature sets and machine learning approaches for the rerankers are evaluated. For the task of opinion expression extraction, the best model shows a 10-point absolute improvement in soft recall on the MPQA corpus over a conventional sequence labeler based on local contextual features, while precision decreases only slightly. Significant improvements are also seen for the extended tasks where holders and polarities are considered: 10 and 7 points in recall, respectively. In addition, the systems outperform previously published results for unlabeled (6 F-measure points) and polarity-labeled (10-15 points) opinion expression extraction. Finally, as an extrinsic evaluation, the extracted MPQA-style opinion expressions are used in practical opinion mining tasks. In all scenarios considered, the machine learning features derived from the opinion expressions lead to statistically significant improvements.",
author = "Richard Johansson and Alessandro Moschitti",
year = "2013",
month = "9",
day = "1",
doi = "10.1162/COLI_a_00141",
language = "English",
volume = "39",
pages = "473--509",
journal = "Computational Linguistics",
issn = "0891-2017",
publisher = "MIT Press Journals",
number = "3",

}

TY - JOUR

T1 - Relational features in fine-grained opinion analysis

AU - Johansson, Richard

AU - Moschitti, Alessandro

PY - 2013/9/1

Y1 - 2013/9/1

N2 - Fine-grained opinion analysis methods often make use of linguistic features but typically do not take the interaction between opinions into account. This article describes a set of experiments that demonstrate that relational features, mainly derived from dependency-syntactic and semantic role structures, can significantly improve the performance of automatic systems for a number of fine-grained opinion analysis tasks: marking up opinion expressions, finding opinion holders, and determining the polarities of opinion expressions. These features make it possible to model the way opinions expressed in natural-language discourse interact in a sentence over arbitrary distances. The use of relations requires us to consider multiple opinions simultaneously, which makes the search for the optimal analysis intractable. However, a reranker can be used as a sufficiently accurate and efficient approximation. A number of feature sets and machine learning approaches for the rerankers are evaluated. For the task of opinion expression extraction, the best model shows a 10-point absolute improvement in soft recall on the MPQA corpus over a conventional sequence labeler based on local contextual features, while precision decreases only slightly. Significant improvements are also seen for the extended tasks where holders and polarities are considered: 10 and 7 points in recall, respectively. In addition, the systems outperform previously published results for unlabeled (6 F-measure points) and polarity-labeled (10-15 points) opinion expression extraction. Finally, as an extrinsic evaluation, the extracted MPQA-style opinion expressions are used in practical opinion mining tasks. In all scenarios considered, the machine learning features derived from the opinion expressions lead to statistically significant improvements.

AB - Fine-grained opinion analysis methods often make use of linguistic features but typically do not take the interaction between opinions into account. This article describes a set of experiments that demonstrate that relational features, mainly derived from dependency-syntactic and semantic role structures, can significantly improve the performance of automatic systems for a number of fine-grained opinion analysis tasks: marking up opinion expressions, finding opinion holders, and determining the polarities of opinion expressions. These features make it possible to model the way opinions expressed in natural-language discourse interact in a sentence over arbitrary distances. The use of relations requires us to consider multiple opinions simultaneously, which makes the search for the optimal analysis intractable. However, a reranker can be used as a sufficiently accurate and efficient approximation. A number of feature sets and machine learning approaches for the rerankers are evaluated. For the task of opinion expression extraction, the best model shows a 10-point absolute improvement in soft recall on the MPQA corpus over a conventional sequence labeler based on local contextual features, while precision decreases only slightly. Significant improvements are also seen for the extended tasks where holders and polarities are considered: 10 and 7 points in recall, respectively. In addition, the systems outperform previously published results for unlabeled (6 F-measure points) and polarity-labeled (10-15 points) opinion expression extraction. Finally, as an extrinsic evaluation, the extracted MPQA-style opinion expressions are used in practical opinion mining tasks. In all scenarios considered, the machine learning features derived from the opinion expressions lead to statistically significant improvements.

UR - http://www.scopus.com/inward/record.url?scp=84881192667&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84881192667&partnerID=8YFLogxK

U2 - 10.1162/COLI_a_00141

DO - 10.1162/COLI_a_00141

M3 - Article

VL - 39

SP - 473

EP - 509

JO - Computational Linguistics

JF - Computational Linguistics

SN - 0891-2017

IS - 3

ER -