An exploratory study on Latent-Dirichlet Allocation models for aspect identification on short sentences

Ameer Abu Bakar; Lay-Ki Soon; Hui-Ngo Goh

doi:10.1007/978-981-10-8276-4_30

An exploratory study on Latent-Dirichlet Allocation models for aspect identification on short sentences

Ameer Abu Bakar, Lay-Ki Soon, Hui-Ngo Goh

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

1 Citation (Scopus)

Abstract

This paper reports an exploratory study conducted to investigate the performance of topic modelling algorithms in aspect identification. Aspect identification is an important step in aspect-based sentiment analysis. Latent-Dirichlet Allocation model serves as the baseline of topic models in the experiments. One of the variations of LDA, namely Phrase-LDA was experimented to benchmark its performance against the original LDA. Although it was reported that PLDA performs better compared to LDA in aspect-based sentiment analysis, our experimental results indicate that LDA works better on dataset with short sentences. A new PLDA model was also proposed by using different types of dependencies to extract the phrases.

Original language	English
Title of host publication	Computational Science and Technology
Subtitle of host publication	4th ICCST 2017, Kuala Lumpur, Malaysia, 29–30 November, 2017
Editors	Rayner Alfred, Hiroyuki Iida, Ag. Asri Ag. Ibrahim, Yuto Lim
Place of Publication	Singapore Singapore
Publisher	Springer
Pages	314-323
Number of pages	10
ISBN (Electronic)	9789811082764
ISBN (Print)	9789811082757
DOIs	https://doi.org/10.1007/978-981-10-8276-4_30
Publication status	Published - 2018
Externally published	Yes
Event	International Conference on Computational Science and Technology 2017 - Kuala Lumpur, Malaysia Duration: 29 Nov 2017 → 30 Nov 2017 Conference number: 4th https://link.springer.com/book/10.1007/978-981-10-8276-4 (Proceedings)

Publication series

Name	Lecture Notes in Electrical Engineering
Publisher	Springer
Volume	488
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	International Conference on Computational Science and Technology 2017
Abbreviated title	ICCST 2017
Country/Territory	Malaysia
City	Kuala Lumpur
Period	29/11/17 → 30/11/17
Internet address	https://link.springer.com/book/10.1007/978-981-10-8276-4 (Proceedings)

Keywords

Aspect identification
Latent-Dirichlet Allocation
Sentiment analysis
Topic modeling

Access to Document

10.1007/978-981-10-8276-4_30

Cite this

Bakar, A. A., Soon, L.-K., & Goh, H.-N. (2018). An exploratory study on Latent-Dirichlet Allocation models for aspect identification on short sentences. In R. Alfred, H. Iida, A. A. Ag. Ibrahim, & Y. Lim (Eds.), Computational Science and Technology: 4th ICCST 2017, Kuala Lumpur, Malaysia, 29–30 November, 2017 (pp. 314-323). (Lecture Notes in Electrical Engineering; Vol. 488). Springer. https://doi.org/10.1007/978-981-10-8276-4_30

Bakar, Ameer Abu ; Soon, Lay-Ki ; Goh, Hui-Ngo. / An exploratory study on Latent-Dirichlet Allocation models for aspect identification on short sentences. Computational Science and Technology: 4th ICCST 2017, Kuala Lumpur, Malaysia, 29–30 November, 2017. editor / Rayner Alfred ; Hiroyuki Iida ; Ag. Asri Ag. Ibrahim ; Yuto Lim. Singapore Singapore : Springer, 2018. pp. 314-323 (Lecture Notes in Electrical Engineering).

@inproceedings{8d06941c030049a08abdacf3084c91ec,

title = "An exploratory study on Latent-Dirichlet Allocation models for aspect identification on short sentences",

abstract = "This paper reports an exploratory study conducted to investigate the performance of topic modelling algorithms in aspect identification. Aspect identification is an important step in aspect-based sentiment analysis. Latent-Dirichlet Allocation model serves as the baseline of topic models in the experiments. One of the variations of LDA, namely Phrase-LDA was experimented to benchmark its performance against the original LDA. Although it was reported that PLDA performs better compared to LDA in aspect-based sentiment analysis, our experimental results indicate that LDA works better on dataset with short sentences. A new PLDA model was also proposed by using different types of dependencies to extract the phrases.",

keywords = "Aspect identification, Latent-Dirichlet Allocation, Sentiment analysis, Topic modeling",

author = "Bakar, {Ameer Abu} and Lay-Ki Soon and Hui-Ngo Goh",

note = "Publisher Copyright: {\textcopyright} 2018, Springer Nature Singapore Pte Ltd. Copyright: Copyright 2018 Elsevier B.V., All rights reserved.; International Conference on Computational Science and Technology 2017, ICCST 2017 ; Conference date: 29-11-2017 Through 30-11-2017",

year = "2018",

doi = "10.1007/978-981-10-8276-4_30",

language = "English",

isbn = "9789811082757",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer",

pages = "314--323",

editor = "Rayner Alfred and Hiroyuki Iida and {Ag. Ibrahim}, {Ag. Asri} and Yuto Lim",

booktitle = "Computational Science and Technology",

url = "https://link.springer.com/book/10.1007/978-981-10-8276-4",

}

Bakar, AA, Soon, L-K & Goh, H-N 2018, An exploratory study on Latent-Dirichlet Allocation models for aspect identification on short sentences. in R Alfred, H Iida, AA Ag. Ibrahim & Y Lim (eds), Computational Science and Technology: 4th ICCST 2017, Kuala Lumpur, Malaysia, 29–30 November, 2017. Lecture Notes in Electrical Engineering, vol. 488, Springer, Singapore Singapore , pp. 314-323, International Conference on Computational Science and Technology 2017, Kuala Lumpur, Malaysia, 29/11/17. https://doi.org/10.1007/978-981-10-8276-4_30

An exploratory study on Latent-Dirichlet Allocation models for aspect identification on short sentences. / Bakar, Ameer Abu; Soon, Lay-Ki; Goh, Hui-Ngo.
Computational Science and Technology: 4th ICCST 2017, Kuala Lumpur, Malaysia, 29–30 November, 2017. ed. / Rayner Alfred; Hiroyuki Iida; Ag. Asri Ag. Ibrahim; Yuto Lim. Singapore Singapore : Springer, 2018. p. 314-323 (Lecture Notes in Electrical Engineering; Vol. 488).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - An exploratory study on Latent-Dirichlet Allocation models for aspect identification on short sentences

AU - Bakar, Ameer Abu

AU - Soon, Lay-Ki

AU - Goh, Hui-Ngo

N1 - Conference code: 4th

PY - 2018

Y1 - 2018

N2 - This paper reports an exploratory study conducted to investigate the performance of topic modelling algorithms in aspect identification. Aspect identification is an important step in aspect-based sentiment analysis. Latent-Dirichlet Allocation model serves as the baseline of topic models in the experiments. One of the variations of LDA, namely Phrase-LDA was experimented to benchmark its performance against the original LDA. Although it was reported that PLDA performs better compared to LDA in aspect-based sentiment analysis, our experimental results indicate that LDA works better on dataset with short sentences. A new PLDA model was also proposed by using different types of dependencies to extract the phrases.

AB - This paper reports an exploratory study conducted to investigate the performance of topic modelling algorithms in aspect identification. Aspect identification is an important step in aspect-based sentiment analysis. Latent-Dirichlet Allocation model serves as the baseline of topic models in the experiments. One of the variations of LDA, namely Phrase-LDA was experimented to benchmark its performance against the original LDA. Although it was reported that PLDA performs better compared to LDA in aspect-based sentiment analysis, our experimental results indicate that LDA works better on dataset with short sentences. A new PLDA model was also proposed by using different types of dependencies to extract the phrases.

KW - Aspect identification

KW - Latent-Dirichlet Allocation

KW - Sentiment analysis

KW - Topic modeling

UR - http://www.scopus.com/inward/record.url?scp=85043342041&partnerID=8YFLogxK

U2 - 10.1007/978-981-10-8276-4_30

DO - 10.1007/978-981-10-8276-4_30

M3 - Conference Paper

AN - SCOPUS:85043342041

SN - 9789811082757

T3 - Lecture Notes in Electrical Engineering

SP - 314

EP - 323

BT - Computational Science and Technology

A2 - Alfred, Rayner

A2 - Iida, Hiroyuki

A2 - Ag. Ibrahim, Ag. Asri

A2 - Lim, Yuto

PB - Springer

CY - Singapore Singapore

T2 - International Conference on Computational Science and Technology 2017

Y2 - 29 November 2017 through 30 November 2017

ER -

Bakar AA, Soon LK, Goh HN. An exploratory study on Latent-Dirichlet Allocation models for aspect identification on short sentences. In Alfred R, Iida H, Ag. Ibrahim AA, Lim Y, editors, Computational Science and Technology: 4th ICCST 2017, Kuala Lumpur, Malaysia, 29–30 November, 2017. Singapore Singapore : Springer. 2018. p. 314-323. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-981-10-8276-4_30

An exploratory study on Latent-Dirichlet Allocation models for aspect identification on short sentences

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this