Home Appliance Review Analysis Via Adversarial Reptile

Tai Jung Kan; Chia Hui Chang

doi:10.1145/3486622.3493958

Home Appliance Review Analysis Via Adversarial Reptile

Tai Jung Kan, Chia Hui Chang

Department of Computer Science and Information Engineering

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

Abstract

Studying discussion of products on social media can help manufacturers improve their products. Opinions provided through online reviews can immediately reflect whether the product is accepted by people, and which aspects of the product are most discussed. In this article, we divide the analysis of home appliances into three tasks, including named entity recognition (NER), aspect category extraction (ACE), and aspect category sentiment classification (ACSC). To improve the performance of ACSC, we combine the Reptile algorithm in meta learning with the concept of domain adversarial training to form the concept of the Adversarial Reptile algorithm. We found that the macro-F1 is improved from 68.6% (BERT fine-tuned model) to 70.3% (p-value 0.04).

Original language	English
Title of host publication	Proceedings - 2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, WI-IAT 2021
Publisher	Association for Computing Machinery
Pages	64-70
Number of pages	7
ISBN (Electronic)	9781450391153
DOIs	https://doi.org/10.1145/3486622.3493958
State	Published - 14 Dec 2021
Event	2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, WI-IAT 2021 - Virtual, Online, Australia Duration: 14 Dec 2021 → 17 Dec 2021

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, WI-IAT 2021
Country/Territory	Australia
City	Virtual, Online
Period	14/12/21 → 17/12/21

Keywords

Meta Learning
Sentiment Analysis
Transfer Learning
adversarial Training

Access to Document

10.1145/3486622.3493958

Cite this

@inproceedings{782e47bc922e41f5b54c59bae6f67051,

title = "Home Appliance Review Analysis Via Adversarial Reptile",

abstract = "Studying discussion of products on social media can help manufacturers improve their products. Opinions provided through online reviews can immediately reflect whether the product is accepted by people, and which aspects of the product are most discussed. In this article, we divide the analysis of home appliances into three tasks, including named entity recognition (NER), aspect category extraction (ACE), and aspect category sentiment classification (ACSC). To improve the performance of ACSC, we combine the Reptile algorithm in meta learning with the concept of domain adversarial training to form the concept of the Adversarial Reptile algorithm. We found that the macro-F1 is improved from 68.6% (BERT fine-tuned model) to 70.3% (p-value 0.04).",

keywords = "Meta Learning, Sentiment Analysis, Transfer Learning, adversarial Training",

author = "Kan, {Tai Jung} and Chang, {Chia Hui}",

note = "Publisher Copyright: {\textcopyright} 2021 ACM.; 2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, WI-IAT 2021 ; Conference date: 14-12-2021 Through 17-12-2021",

year = "2021",

month = dec,

day = "14",

doi = "10.1145/3486622.3493958",

language = "???core.languages.en_GB???",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "64--70",

booktitle = "Proceedings - 2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, WI-IAT 2021",

}

Kan, TJ & Chang, CH 2021, Home Appliance Review Analysis Via Adversarial Reptile. in Proceedings - 2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, WI-IAT 2021. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 64-70, 2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, WI-IAT 2021, Virtual, Online, Australia, 14/12/21. https://doi.org/10.1145/3486622.3493958

Home Appliance Review Analysis Via Adversarial Reptile. / Kan, Tai Jung; Chang, Chia Hui.
Proceedings - 2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, WI-IAT 2021. Association for Computing Machinery, 2021. p. 64-70 (ACM International Conference Proceeding Series).

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

TY - GEN

T1 - Home Appliance Review Analysis Via Adversarial Reptile

AU - Kan, Tai Jung

AU - Chang, Chia Hui

PY - 2021/12/14

Y1 - 2021/12/14

N2 - Studying discussion of products on social media can help manufacturers improve their products. Opinions provided through online reviews can immediately reflect whether the product is accepted by people, and which aspects of the product are most discussed. In this article, we divide the analysis of home appliances into three tasks, including named entity recognition (NER), aspect category extraction (ACE), and aspect category sentiment classification (ACSC). To improve the performance of ACSC, we combine the Reptile algorithm in meta learning with the concept of domain adversarial training to form the concept of the Adversarial Reptile algorithm. We found that the macro-F1 is improved from 68.6% (BERT fine-tuned model) to 70.3% (p-value 0.04).

AB - Studying discussion of products on social media can help manufacturers improve their products. Opinions provided through online reviews can immediately reflect whether the product is accepted by people, and which aspects of the product are most discussed. In this article, we divide the analysis of home appliances into three tasks, including named entity recognition (NER), aspect category extraction (ACE), and aspect category sentiment classification (ACSC). To improve the performance of ACSC, we combine the Reptile algorithm in meta learning with the concept of domain adversarial training to form the concept of the Adversarial Reptile algorithm. We found that the macro-F1 is improved from 68.6% (BERT fine-tuned model) to 70.3% (p-value 0.04).

KW - Meta Learning

KW - Sentiment Analysis

KW - Transfer Learning

KW - adversarial Training

UR - http://www.scopus.com/inward/record.url?scp=85128607906&partnerID=8YFLogxK

U2 - 10.1145/3486622.3493958

DO - 10.1145/3486622.3493958

M3 - 會議論文篇章

AN - SCOPUS:85128607906

T3 - ACM International Conference Proceeding Series

SP - 64

EP - 70

BT - Proceedings - 2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, WI-IAT 2021

PB - Association for Computing Machinery

T2 - 2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology, WI-IAT 2021

Y2 - 14 December 2021 through 17 December 2021

ER -

Home Appliance Review Analysis Via Adversarial Reptile

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this