Learning Sentiment Based Ranked-Lexicons for Opinion Retrieval

Filipa Peleja, João Magalhães

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In contrast to classic search where users look for factual information, opinion retrieval aims at finding and ranking subjective information. A major challenge of opinion retrieval is the informal nature of user reviews and the domain specific jargon used to describe the targeted item. In this paper, we present an automatic method to learn a space model for opinion retrieval. Our approach is a generative model that learns sentiment word distributions by embedding multi-level relevance judgments in the estimation of the model parameters. In addition to sentiment word distributions, we also infer domain specific named entities that due to their popularity become a sentiment reference in their domain (e.g. name of a movie, “Batman” or specific hotel items, “carpet”). This contrasts with previous approaches that learn a word’s polarity or aspect-based polarity. Opinion retrieval experiments were done in two large datasets with over 703.000 movie reviews and 189.000 hotel reviews. The proposed method achieved better, or equal, performance than the benchmark baselines.
Original languageEnglish
Title of host publicationAdvances in Information Retrieval - 37th European Conference on IR Research, ECIR 2015, Vienna, Austria, March 29 - April 2, 2015. Proceedings
PublisherSpringer-Verlag
Pages435-440
Number of pages6
Volume9022
ISBN (Electronic)978-331916353-6
DOIs
Publication statusPublished - 2015
Event37th European Conference on IR Research, : ECIR 2015 - Vienna, Vienna, Austria
Duration: 29 Mar 20152 Apr 2015

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Verlag
Volume9022
ISSN (Print)0302-9743

Conference

Conference37th European Conference on IR Research,
Abbreviated titleECIR 2015
CountryAustria
CityVienna
Period29/03/152/04/15

Fingerprint

Hotels
Information retrieval
Experiments

Keywords

  • Opinion retrieval
  • Sentiment analysis
  • Sentiment space model

Cite this

Peleja, F., & Magalhães, J. (2015). Learning Sentiment Based Ranked-Lexicons for Opinion Retrieval. In Advances in Information Retrieval - 37th European Conference on IR Research, ECIR 2015, Vienna, Austria, March 29 - April 2, 2015. Proceedings (Vol. 9022, pp. 435-440). (Lecture Notes in Computer Science; Vol. 9022). Springer-Verlag. https://doi.org/10.1007/978-3-319-16354-3_47
Peleja, Filipa ; Magalhães, João. / Learning Sentiment Based Ranked-Lexicons for Opinion Retrieval. Advances in Information Retrieval - 37th European Conference on IR Research, ECIR 2015, Vienna, Austria, March 29 - April 2, 2015. Proceedings. Vol. 9022 Springer-Verlag, 2015. pp. 435-440 (Lecture Notes in Computer Science).
@inproceedings{1630b7e9612a4fad8f9aa8b7e9ba9e6c,
title = "Learning Sentiment Based Ranked-Lexicons for Opinion Retrieval",
abstract = "In contrast to classic search where users look for factual information, opinion retrieval aims at finding and ranking subjective information. A major challenge of opinion retrieval is the informal nature of user reviews and the domain specific jargon used to describe the targeted item. In this paper, we present an automatic method to learn a space model for opinion retrieval. Our approach is a generative model that learns sentiment word distributions by embedding multi-level relevance judgments in the estimation of the model parameters. In addition to sentiment word distributions, we also infer domain specific named entities that due to their popularity become a sentiment reference in their domain (e.g. name of a movie, “Batman” or specific hotel items, “carpet”). This contrasts with previous approaches that learn a word’s polarity or aspect-based polarity. Opinion retrieval experiments were done in two large datasets with over 703.000 movie reviews and 189.000 hotel reviews. The proposed method achieved better, or equal, performance than the benchmark baselines.",
keywords = "Opinion retrieval, Sentiment analysis, Sentiment space model",
author = "Filipa Peleja and Jo{\~a}o Magalh{\~a}es",
note = "Sem PDF.",
year = "2015",
doi = "10.1007/978-3-319-16354-3_47",
language = "English",
volume = "9022",
series = "Lecture Notes in Computer Science",
publisher = "Springer-Verlag",
pages = "435--440",
booktitle = "Advances in Information Retrieval - 37th European Conference on IR Research, ECIR 2015, Vienna, Austria, March 29 - April 2, 2015. Proceedings",

}

Peleja, F & Magalhães, J 2015, Learning Sentiment Based Ranked-Lexicons for Opinion Retrieval. in Advances in Information Retrieval - 37th European Conference on IR Research, ECIR 2015, Vienna, Austria, March 29 - April 2, 2015. Proceedings. vol. 9022, Lecture Notes in Computer Science, vol. 9022, Springer-Verlag, pp. 435-440, 37th European Conference on IR Research, , Vienna, Austria, 29/03/15. https://doi.org/10.1007/978-3-319-16354-3_47

Learning Sentiment Based Ranked-Lexicons for Opinion Retrieval. / Peleja, Filipa; Magalhães, João.

Advances in Information Retrieval - 37th European Conference on IR Research, ECIR 2015, Vienna, Austria, March 29 - April 2, 2015. Proceedings. Vol. 9022 Springer-Verlag, 2015. p. 435-440 (Lecture Notes in Computer Science; Vol. 9022).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Learning Sentiment Based Ranked-Lexicons for Opinion Retrieval

AU - Peleja, Filipa

AU - Magalhães, João

N1 - Sem PDF.

PY - 2015

Y1 - 2015

N2 - In contrast to classic search where users look for factual information, opinion retrieval aims at finding and ranking subjective information. A major challenge of opinion retrieval is the informal nature of user reviews and the domain specific jargon used to describe the targeted item. In this paper, we present an automatic method to learn a space model for opinion retrieval. Our approach is a generative model that learns sentiment word distributions by embedding multi-level relevance judgments in the estimation of the model parameters. In addition to sentiment word distributions, we also infer domain specific named entities that due to their popularity become a sentiment reference in their domain (e.g. name of a movie, “Batman” or specific hotel items, “carpet”). This contrasts with previous approaches that learn a word’s polarity or aspect-based polarity. Opinion retrieval experiments were done in two large datasets with over 703.000 movie reviews and 189.000 hotel reviews. The proposed method achieved better, or equal, performance than the benchmark baselines.

AB - In contrast to classic search where users look for factual information, opinion retrieval aims at finding and ranking subjective information. A major challenge of opinion retrieval is the informal nature of user reviews and the domain specific jargon used to describe the targeted item. In this paper, we present an automatic method to learn a space model for opinion retrieval. Our approach is a generative model that learns sentiment word distributions by embedding multi-level relevance judgments in the estimation of the model parameters. In addition to sentiment word distributions, we also infer domain specific named entities that due to their popularity become a sentiment reference in their domain (e.g. name of a movie, “Batman” or specific hotel items, “carpet”). This contrasts with previous approaches that learn a word’s polarity or aspect-based polarity. Opinion retrieval experiments were done in two large datasets with over 703.000 movie reviews and 189.000 hotel reviews. The proposed method achieved better, or equal, performance than the benchmark baselines.

KW - Opinion retrieval

KW - Sentiment analysis

KW - Sentiment space model

U2 - 10.1007/978-3-319-16354-3_47

DO - 10.1007/978-3-319-16354-3_47

M3 - Conference contribution

VL - 9022

T3 - Lecture Notes in Computer Science

SP - 435

EP - 440

BT - Advances in Information Retrieval - 37th European Conference on IR Research, ECIR 2015, Vienna, Austria, March 29 - April 2, 2015. Proceedings

PB - Springer-Verlag

ER -

Peleja F, Magalhães J. Learning Sentiment Based Ranked-Lexicons for Opinion Retrieval. In Advances in Information Retrieval - 37th European Conference on IR Research, ECIR 2015, Vienna, Austria, March 29 - April 2, 2015. Proceedings. Vol. 9022. Springer-Verlag. 2015. p. 435-440. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-319-16354-3_47